{"behavior":"degenerate_repetition","label_filter":null,"limit":10,"next_offset":52543,"offset":52533,"previous_offset":null,"rows":[{"assistant_initiates":false,"asst_text":"G\u0142\u00f3wnym problemem, z kt\u00f3rym si\u0119 borykasz, wydaje si\u0119 by\u0107 brak odpowiedniego ustawienia instrument\u00f3w wirtualnych (VSTi) lub generator\u00f3w d\u017awi\u0119ku w programach DAW takich jak Reaper i Studio One. MIDI samo w sobie nie zawiera d\u017awi\u0119k\u00f3w, a jedynie informacje o nutach, ich d\u0142ugo\u015bci, dynamice itp. Aby te informacje przekszta\u0142ci\u0107 w rzeczywiste d\u017awi\u0119ki, potrzebujesz odpowiedniego instrumentu wirtualnego lub syntezatora.\n\nOto kilka krok\u00f3w, kt\u00f3re mog\u0105 pom\u00f3c rozwi\u0105za\u0107 problem:\n\n1. **Sprawd\u017a \u015acie\u017ck\u0119 MIDI:** Upewnij si\u0119, \u017ce \u015bcie\u017cka MIDI w twoim projekcie Reaper lub Studio One ma przypisany odpowiedni instrument VSTi.\n    - W Reaper, za\u0142\u00f3\u017c now\u0105 \u015bcie\u017ck\u0119 (Track -> Insert new track) i przeci\u0105gnij sw\u00f3j plik MIDI na t\u0119 \u015bcie\u017ck\u0119.\n    - Kliknij przycisk FX na \u015bcie\u017cce, aby doda\u0107 instrument wirtualny (np. VSTi). Wybierz instrument, kt\u00f3ry chcesz u\u017cywa\u0107 do odtwarzania pliku MIDI i dodaj go.\n\n2. **Za\u0142aduj Instrument Wirtualny:** Po przypisaniu \u015bcie\u017cki MIDI do instrumentu wirtualnego, za\u0142aduj odpowiednie brzmienie/sampl w instrumencie. \n    - W Studio One musisz przypisa\u0107 odpowiedni instrument do \u015bcie\u017cki MIDI, podobnie jak w Reaper.\n\n3. **Sprawd\u017a Ustawienia Audio:** Upewnij si\u0119, \u017ce wyj\u015bcie audio w DAW (Reaper, Studio One) jest poprawnie skonfigurowane i dzia\u0142a przez ASIO. \n    - W Reaper, przejd\u017a do Options -> Preferences -> Audio -> Device i upewnij si\u0119, \u017ce Twoje urz\u0105dzenie ASIO jest wybrane. \n    - W Studio One, przejd\u017a do Studio One -> Options -> Audio Setup i upewnij si\u0119, \u017ce urz\u0105dzenie ASIO jest poprawnie wybrane.\n\n4. **Sprawd\u017a Ustawienia MIDI:** Upewnij si\u0119, \u017ce dane MIDI s\u0105 wysy\u0142ane do odpowiedniego instrumentu.\n    - Upewnij si\u0119, \u017ce \u015bcie\u017cka MIDI jest przypisana do w\u0142a\u015bciwego kana\u0142u MIDI i \u017ce instrument wirtualny jest ustawiony na odbieranie danych z tego samego kana\u0142u.\n\n5. **Plik MIDI:** Mo\u017cliwe, \u017ce plik MIDI jest przypisany do specyficznych brzmie\u0144/kana\u0142\u00f3w, kt\u00f3re nie s\u0105 obecnie skonfigurowane w Twoim projekcie DAW. Upewnij si\u0119, \u017ce plik MIDI jest mapowany prawid\u0142owo do instrument\u00f3w, kt\u00f3re masz dost\u0119pne.\n\n6. **Instrumenty General MIDI:** Je\u017celi korzystasz z prostych instrument\u00f3w GM (General MIDI), mo\u017cesz spr\u00f3bowa\u0107 u\u017cy\u0107 wirtualnego syntezatora General MIDI, takiego jak \"Microsoft GS Wavetable Synth\" w narz\u0119dziu DAW.\n\nSprawd\u017a te kroki krok po kroku i powiedz, czy kt\u00f3ry\u015b z nich rozwi\u0105za\u0142 Tw\u00f3j problem. W przypadku dalszych trudno\u015bci, dostarcz wi\u0119cej szczeg\u00f3\u0142\u00f3w, aby mo\u017cna by\u0142o bardziej precyzyjnie zdiagnozowa\u0107 problem.","behavior":"degenerate_repetition","conversation_hash":"6a2c7d9bd1a1db5fd3265740f598662d","conversation_redacted":false,"conversation_toxic":false,"country":"Poland","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Polish","model":"gpt-4o-2024-05-13","n_chars":2907,"n_turns_in_conversation":2,"n_words":438,"source_row_offset":52533,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:33:32 GMT","turn_id":"95df670da7321fa4a951eef9","turn_index":0,"user_text":"Dzi\u0119ki odpali\u0142em to ASIO - super dzia\u0142a.\nTeraz mam tylko problem, \u017ce jak odpali\u0142em np. Reaper i otworzy\u0142em plik midi - to nic nie s\u0142ysz\u0119, wygl\u0105da to tak jakby baza sampli (d\u017awi\u0119k\u00f3w by\u0142a nie za\u0142adowana). To samo si\u0119 dzia\u0142o w Studio One, ale np. jak utworzy\u0142em na szybko sam jaki\u015b rytm to ju\u017c by\u0142o s\u0142ycha\u0107 d\u017awi\u0119ki. \nJak te pliki odpalam normalnie w windows 10 to \u0142adnie s\u0142ycha\u0107 i wszystko jest z nimi w porz\u0105dku.\nCo mo\u017ce by\u0107 przyczyn\u0105 \u017ce w tych programach nie s\u0142ysz\u0119 d\u017awi\u0119k\u00f3w?"},{"assistant_initiates":false,"asst_text":"Wygl\u0105da na to, \u017ce problem jest zwi\u0105zany z brakiem dost\u0119pnych wirtualnych instrument\u00f3w (VSTi) oraz z poprawnym przypisaniem instrument\u00f3w MIDI do danych wej\u015bciowych. Aby rozwi\u0105za\u0107 ten problem, potrzebujesz pobra\u0107 i zainstalowa\u0107 dodatkowe instrumenty wirtualne oraz poprawnie skonfigurowa\u0107 wyj\u015bcia MIDI.\n\n### Kroki do rozwi\u0105zania problemu:\n\n#### 1. Instalacja Dodatkowych Wirtualnych Instrument\u00f3w (VSTi)\nMo\u017cesz pobra\u0107 darmowe wirtualne instrumenty (VSTi), kt\u00f3re b\u0119d\u0105 kompatybilne z Reaperem. Oto kilka propozycji:\n\n- **SFZ Player:**\n  - Pobierz SFZ Player, kt\u00f3ry pozwala na odtwarzanie plik\u00f3w soundfont (.sf2). Mo\u017cesz go znale\u017a\u0107 np. na stronie [Plogue sforzando](https://www.plogue.com/products/sforzando.html).\n  - Po zainstalowaniu SFZ Playera, mo\u017cesz pobra\u0107 darmowe soundfonty (.sf2) z internetu, np. na stronie [MuseScore](https://www.musescore.org/en/soundfont).\n\n- **TX16Wx Software Sampler:**\n  - Pobierz darmowy sampler VSTi, np. TX16Wx: [TX16Wx](http://www.tx16wx.com/).\n\n- **Versilian Studios:**\n  - Pobierz darmowe instrumenty dost\u0119pne na stronie [Versilian Studios](https://vis.versilstudios.com/vsco-community.html).\n\n**Instalacja VSTi:**\n- Po pobraniu instrument\u00f3w, zainstaluj je zgodnie z instrukcjami.\n- Dodaj \u015bcie\u017ck\u0119 do folderu, w kt\u00f3rym masz zainstalowane VSTi, do preferencji w Reaperze (Options -> Preferences -> VST -> VST plug-in paths).\n\n#### 2. Konfiguracja MIDI w Reaperze\nUpewnij si\u0119, \u017ce MIDI jest poprawnie skonfigurowane:\n\n1. **Ustaw Instrument Wirtualny:**\n   - Tworz\u0105c now\u0105 \u015bcie\u017ck\u0119 w Reaperze, dodaj instrument wirtualny na \u015bcie\u017ck\u0119 MIDI, np. SFZ Playera lub TX16Wx.\n   - Importuj odpowiedni plik soundfont (.sf2) lub sample do tego instrumentu.\n\n2. **Microsoft GS Wavetable Synth:**\n   - W Reaper Preferences przejd\u017a do Audio -> MIDI Devices.\n   - Znajd\u017a Microsoft GS Wavetable Synth na li\u015bcie MIDI outputs i upewnij si\u0119, \u017ce jest w\u0142\u0105czony do u\u017cycia (enabled). Je\u015bli \"Not found\" oznacza brak sterownika, by\u0107 mo\u017ce warto spr\u00f3bowa\u0107 uruchomi\u0107 Reaper jako administrator.\n\n#### 3. Testowanie \u015acie\u017cek MIDI\nPo wykonaniu powy\u017cszych krok\u00f3w:\n\n- Za\u0142aduj sw\u00f3j plik MIDI do Reaper.\n- Przypisz ka\u017cd\u0105 \u015bcie\u017ck\u0119 MIDI do odpowiedniego wirtualnego instrumentu.\n- Upewnij si\u0119, \u017ce ka\u017cda \u015bcie\u017cka MIDI wysy\u0142a dane na prawid\u0142owy port i kana\u0142 MIDI (mo\u017ce by\u0107 to widoczne poprzez w\u0142a\u015bciwo\u015bci \u015bcie\u017cki).\n\n#### 4. Monitorowanie i Sprawdzanie D\u017awi\u0119ku \n- Upewnij si\u0119, \u017ce w mikserze Reaper wida\u0107, \u017ce \u015bcie\u017cka MIDI wysy\u0142a sygna\u0142 do instrumentu.\n- Sprawd\u017a, czy na mikserze widzisz aktywno\u015b\u0107 na \u015bcie\u017cce audio odpowiadaj\u0105cej wirtualnemu instrumentowi.\n\n#### Dodatkowe Porady:\nJe\u017celi wci\u0105\u017c masz problemy:\n\n- Spr\u00f3buj zaktualizowa\u0107 sterowniki audio i MIDI dla swojego systemu.\n- Otw\u00f3rz inne programy DAW (Studio One) i wykonaj podobne kroki, aby zobaczy\u0107, czy problem jest bardziej zwi\u0105zany z konfiguracj\u0105 konkretnego DAW.\n\nMam nadziej\u0119, \u017ce te kroki pomog\u0105 Ci zdiagnozowa\u0107 i rozwi\u0105za\u0107 problem!","behavior":"degenerate_repetition","conversation_hash":"6a2c7d9bd1a1db5fd3265740f598662d","conversation_redacted":false,"conversation_toxic":false,"country":"Poland","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Polish","model":"gpt-4o-2024-05-13","n_chars":3500,"n_turns_in_conversation":2,"n_words":480,"source_row_offset":52534,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:33:32 GMT","turn_id":"9c8edb99ef6a778946c84ab3","turn_index":1,"user_text":"Nie dzia\u0142a.\nW dziale VSTi mam tylko trzy virtualne instrumenty.\nReaSamplOmatic5000 (Cockos)\nReaSynDr (Cockos) (4out)\nReaSynth (Cockos)\n\nPr\u00f3bowa\u0142em w Reaper Preference i w MIDI Devices w ramce MIDI outputs mam Status - not found dotycz\u0105ce jedynego wiersza Microsoft GS Wavetable Synth i ID ustawione na 0.\nMo\u017ce w tym jest problem.\n\nA sk\u0105d do reaper'a mam zainstalowa\u0107 jakie\u015b proste standardowe virtualne instrumenty?\n\nJeszcze jedno co mnie dziwi to fakt \u017ce jak puszczam utw\u00f3r midi w reaper to wszystkie \u015bcie\u017cki wygl\u0105daj\u0105 na aktywne i takie \u015bwiate\u0142ko mruga na pomara\u0144czowo - co\u015b a'la spectrum wykazuj\u0105c\u0105 moc granych d\u017awi\u0119k\u00f3w."},{"assistant_initiates":false,"asst_text":"It seems like your message is a bit jumbled. Could you please provide more context or clarify what you meant to ask? That way, I'd be happy to help you out!","behavior":"degenerate_repetition","conversation_hash":"c28bfa7faad446bc9f8fa04b2ed8a6f8","conversation_redacted":false,"conversation_toxic":false,"country":"Russia","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"English","model":"gpt-4o-2024-05-13","n_chars":177,"n_turns_in_conversation":2,"n_words":35,"source_row_offset":52535,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:33:44 GMT","turn_id":"f361a34f042951618f34d5c6","turn_index":0,"user_text":"gjxtve gfyjc jn xfz?"},{"assistant_initiates":false,"asst_text":"That's quite a puzzle you've presented! I'll give it a shot. You wrote:\n\n\"gjxtve gfyjc jn xfz?\"\n\nConsidering the keys on a QWERTY keyboard and possible typos, it might be:\n\n\"Did you enjoy the tea?\"\n\nIf that's not correct, could you provide any hints or further context? That would help me make a better guess!","behavior":"degenerate_repetition","conversation_hash":"c28bfa7faad446bc9f8fa04b2ed8a6f8","conversation_redacted":false,"conversation_toxic":false,"country":"Russia","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"English","model":"gpt-4o-2024-05-13","n_chars":336,"n_turns_in_conversation":2,"n_words":61,"source_row_offset":52536,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:33:44 GMT","turn_id":"5f02b442a6fd529009c5cf84","turn_index":1,"user_text":"try to guess what i wrote "},{"assistant_initiates":false,"asst_text":"In the late 2000s, there were some key shifts as Kids' WB transitioned over to CW4Kids. Keep in mind that the actual programming could vary slightly depending on the region and network decisions. Below are approximations of the schedules based on typical lineups and transitions during those time periods in 2008:\n\n### Kids' WB - May 17, 2008 (Final Weeks)\n\nBy May 17, 2008, Kids' WB was nearing the end of its run on broadcast television, phasing out in preparation for the launch of the CW4Kids. The lineup might have still included some of the following shows:\n\n1. **The Batman** - Animated series focusing on the Dark Knight battling crime in Gotham.\n2. **Shaggy & Scooby-Doo Get a Clue!** - New adventures with the iconic duo Shaggy and Scooby-Doo.\n3. **Tom and Jerry Tales** - Modern tales featuring the classic cat and mouse duo.\n4. **Johnny Test** - Animated series about a boy and his wild experiments with science.\n5. **Legion of Super Heroes** - Based on the DC Comics team in a futuristic setting.\n6. **Spider Riders** - Animated series featuring Earth's defender battling alongside arachnid warriors in the Inner World.\n\n### CW4Kids - May 24, 2008 (Launch)\n\nCW4Kids officially launched on May 24, 2008, presenting new shows and continuing some from the previous Kids' WB programming. The probable lineup included:\n\n1. **Spectacular Spider-Man** - Animated series focusing on a younger Peter Parker.\n2. **Yu-Gi-Oh! GX** - Continuing adventures in the Yu-Gi-Oh! universe focusing on Duel Monsters.\n3. **The Batman**\n4. **Tom and Jerry Tales**\n5. **Johnny Test**\n6. **World of Quest** - Animated series about an epic quest in a fantastical world.\n\n### CW4Kids - July 2008 \n\nBy July 2008, the network may have stabilized into a regular broadcasting pattern, continuing some cartoons from the initial lineup and perhaps introducing new or rotated programs:\n\n1. **Spectacular Spider-Man**\n2. **Yu-Gi-Oh! GX**\n3. **Dinosaur King** - An adventure series involving dinosaurs and cards.\n4. **The Batman**\n5. **Tom and Jerry Tales**\n6. **Chaos** - Potentially premiering new episodes or a new series.\n\n### CW4Kids - November 2008\n\nHeading towards the end of 2008, the lineup might have looked similar to earlier in the year with some holiday specials or marathons possibly included:\n\n1. **Teenage Mutant Ninja Turtles: Fast Forward** - Following the futuristic adventures of the TMNT.\n2. **Yu-Gi-Oh! 5D's** - New series set in a dystopian future focusing on Turbo Duels.\n3. **Spectacular Spider-Man**\n4. **Huntik: Secrets & Seekers** - Animated series about a group of Seekers searching for magical amulets.\n5. **The Batman**\n6. **Skunk Fu!** - Animated series featuring Skunk and his martial arts adventures.\n\nThese schedules are approximate and may not be 100% accurate but provide a good idea of the progression and changes in the programming during 2008 for Kids WB transitioning into CW4Kids.","behavior":"degenerate_repetition","conversation_hash":"fa73b4e406d2f4461e25d87926ca870e","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"English","model":"gpt-4-turbo-2024-04-09","n_chars":3009,"n_turns_in_conversation":1,"n_words":489,"source_row_offset":52537,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:35:32 GMT","turn_id":"e10aeabe3d3fca59027484b8","turn_index":0,"user_text":"Write a approximation of the Kids WB May 17 2008, CW4Kids May 24 2008, July 2008 and November 2008 schedules "},{"assistant_initiates":false,"asst_text":"Certainly! Let's dive into an example of a simple C++ class.\n\nConsider a class `Rectangle` that can be used to represent and operate on rectangular shapes. We'll include data members for the dimensions of the rectangle, namely its width and height. Additionally, we'll provide some member functions to set and get these dimensions, as well as to calculate the area and perimeter of the rectangle.\n\nHere's a simple implementation:\n\n```cpp\n#include <iostream>\nusing namespace std;\n\nclass Rectangle {\nprivate:\n    // Private data members\n    double width;\n    double height;\n\npublic:\n    // Constructor\n    Rectangle(double w, double h) : width(w), height(h) {}\n\n    // Default constructor\n    Rectangle() : width(0), height(0) {}\n\n    // Setter for width\n    void setWidth(double w) {\n        if (w >= 0) {\n            width = w;\n        } else {\n            cout << \"Width cannot be negative!\" << endl;\n        }\n    }\n\n    // Setter for height\n    void setHeight(double h) {\n        if (h >= 0) {\n            height = h;\n        } else {\n            cout << \"Height cannot be negative!\" << endl;\n        }\n    }\n\n    // Getter for width\n    double getWidth() const {\n        return width;\n    }\n\n    // Getter for height\n    double getHeight() const {\n        return height;\n    }\n\n    // Function to calculate area\n    double area() const {\n        return width * height;\n    }\n\n    // Function to calculate perimeter\n    double perimeter() const {\n        return 2 * (width + height);\n    }\n};\n\nint main() {\n    // Create a Rectangle object using the parameterized constructor\n    Rectangle rect1(10, 5);\n\n    // Use member functions to access data and perform calculations\n    cout << \"Width: \" << rect1.getWidth() << endl;\n    cout << \"Height: \" << rect1.getHeight() << endl;\n    cout << \"Area: \" << rect1.area() << endl;\n    cout << \"Perimeter: \" << rect1.perimeter() << endl;\n\n    // Create a Rectangle object using the default constructor\n    Rectangle rect2;\n\n    // Set the dimensions using setters\n    rect2.setWidth(7);\n    rect2.setHeight(3);\n\n    cout << \"Width: \" << rect2.getWidth() << endl;\n    cout << \"Height: \" << rect2.getHeight() << endl;\n    cout << \"Area: \" << rect2.area() << endl;\n    cout << \"Perimeter: \" << rect2.perimeter() << endl;\n\n    return 0;\n}\n```\n\n### Explanation\n\n1. **Data Members**:\n    - `double width;`\n    - `double height;`\n    These are private data members that store the width and height of the rectangle. By making them private, we encapsulate the data and restrict direct access from outside the class.\n\n2. **Constructors**:\n    - `Rectangle(double w, double h) : width(w), height(h) {}`\n      This is a parameterized constructor that initializes the width and height with given values.\n    - `Rectangle() : width(0), height(0) {}`\n      This is a default constructor that initializes the width and height to 0.\n\n3. **Setter and Getter Methods**:\n    - Setter methods (`setWidth` and `setHeight`) allow controlled modification of the rectangle's dimensions. They include checks to ensure the width and height are non-negative.\n    - Getter methods (`getWidth` and `getHeight`) provide access to the rectangle's dimensions.\n\n4. **Member Functions**:\n    - `double area() const` calculates and returns the area of the rectangle.\n    - `double perimeter() const` calculates and returns the perimeter of the rectangle.\n\n5. **Main Function**:\n    - Demonstrates creating `Rectangle` objects using both constructors.\n    - Manipulates rectangle dimensions using setters.\n    - Retrieves and prints dimensions, area, and perimeter using getters and other member functions.\n\nThis example illustrates basic principles of object-oriented programming in C++: encapsulation, constructors, member functions, and how to work with objects.","behavior":"degenerate_repetition","conversation_hash":"2fb9f20e8c75fd65d276537ca2f73995","conversation_redacted":false,"conversation_toxic":false,"country":"France","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"English","model":"gpt-4o-2024-05-13","n_chars":3814,"n_turns_in_conversation":1,"n_words":541,"source_row_offset":52538,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:36:16 GMT","turn_id":"96fcd15462642a6414318262","turn_index":0,"user_text":"write an example of C++ class and explain"},{"assistant_initiates":false,"asst_text":"Si un delito ocurre en 109 Salinas Ct, Pittsburg, California 94565, varios organismos pueden intervenir en funci\u00f3n de la naturaleza y la gravedad del delito. Aqu\u00ed tienes una lista de las principales agencias que podr\u00edan estar involucradas:\n\n1. **Departamento de Polic\u00eda de Pittsburg**: Esta es la agencia de aplicaci\u00f3n de la ley local que tendr\u00eda la jurisdicci\u00f3n primaria sobre delitos que ocurren dentro de los l\u00edmites de la ciudad de Pittsburg. Ellos ser\u00edan los primeros en responder a la escena, realizar\u00edan la investigaci\u00f3n inicial y tomar\u00edan acci\u00f3n inmediata.\n\n2. **Oficina del Sheriff del Condado de Contra Costa**: Si un delito es particularmente grave o si es necesario apoyo adicional, la oficina del Sheriff podr\u00eda intervenir. Tambi\u00e9n tienen la autoridad para actuar en \u00e1reas no incorporadas dentro del condado.\n\n3. **California Highway Patrol (CHP)**: Si el delito involucra carreteras, autopistas u ocurre en \u00e1reas bajo su jurisdicci\u00f3n particular, el CHP podr\u00eda estar involucrado.\n\n4. **Fiscal de Distrito del Condado de Contra Costa**: Esta oficina se encarga de la prosecuci\u00f3n de delitos en el Condado de Contra Costa. Si el departamento de polic\u00eda local u otra agencia presenta cargos, el fiscal de distrito se encargar\u00e1 de llevar el caso a la corte.\n\n5. **Agencias Federales**: Para ciertos tipos de delitos como terrorismo, fraude financiero, o tr\u00e1fico de drogas a gran escala, agencias federales como el FBI, DEA o ATF podr\u00edan involucrarse.\n\n6. **Servicios de Emergencia M\u00e9dica**: Si hay lesiones involucradas, los servicios de emergencias m\u00e9dicas locales tambi\u00e9n responder\u00edan a la escena.\n\nLa cooperaci\u00f3n entre estas agencias depender\u00e1 del tipo de delito y los recursos necesarios para abordarlo.","behavior":"degenerate_repetition","conversation_hash":"9ba422e6dc1fbfbdc077b864be421fe3","conversation_redacted":false,"conversation_toxic":false,"country":"Argentina","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Spanish","model":"gpt-4o-2024-05-13","n_chars":1799,"n_turns_in_conversation":2,"n_words":278,"source_row_offset":52539,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:36:48 GMT","turn_id":"9249ac3dd7a75bc217554542","turn_index":0,"user_text":"si un delito ocurre en 109 Salinas Ct, Pittsburg, California 94565 quien interviene"},{"assistant_initiates":false,"asst_text":"El c\u00f3digo de \u00e1rea 832 pertenece al estado de Texas en los Estados Unidos. Espec\u00edficamente, cubre la regi\u00f3n metropolitana de Houston, junto con los c\u00f3digos de \u00e1rea 713, 281 y 346. Estos c\u00f3digos de \u00e1rea se superponen en la misma regi\u00f3n geogr\u00e1fica, lo que significa que el \u00e1rea de Houston tiene varios c\u00f3digos de \u00e1rea en uso simult\u00e1neamente.","behavior":"degenerate_repetition","conversation_hash":"9ba422e6dc1fbfbdc077b864be421fe3","conversation_redacted":false,"conversation_toxic":false,"country":"Argentina","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Spanish","model":"gpt-4o-2024-05-13","n_chars":361,"n_turns_in_conversation":2,"n_words":63,"source_row_offset":52540,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:36:48 GMT","turn_id":"96553af1bb31e58026644a33","turn_index":1,"user_text":"(832) de que estado es"},{"assistant_initiates":false,"asst_text":"Merhaba! Size nas\u0131l yard\u0131mc\u0131 olabilirim?","behavior":"degenerate_repetition","conversation_hash":"dca5a75ef8e2b4a040275983ec758535","conversation_redacted":true,"conversation_toxic":false,"country":"T\u00fcrkiye","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Turkish","model":"gpt-4-turbo-2024-04-09","n_chars":48,"n_turns_in_conversation":3,"n_words":6,"source_row_offset":52541,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:37:23 GMT","turn_id":"772326611837cee2afddb062","turn_index":0,"user_text":"merhaba"},{"assistant_initiates":false,"asst_text":"Bu metin, m\u00fcltecilerin ulusal e\u011fitim sistemlerine dahil edilmesiyle ilgili zorluklar\u0131 ve \u00e7\u00f6z\u00fcm \u00f6nerilerini tart\u0131\u015fan olduk\u00e7a kapsaml\u0131 bir literat\u00fcr taramas\u0131 ve analiz i\u00e7ermektedir. Sarah Dryden-Peterson ve ekibinin 2018 tarihli bu \u00e7al\u0131\u015fmas\u0131, \u00f6zellikle Uganda, Kenya ve L\u00fcbnan'daki m\u00fclteci e\u011fitimi uygulamalar\u0131n\u0131 incelemektedir. Metin, m\u00fcltecilerin e\u011fitim hakk\u0131n\u0131 temel almakta ve onlar\u0131n e\u011fitim sistemlerine entegrasyonunu \"yap\u0131sal entegrasyon\" ve \"ili\u015fkisel entegrasyon\" kavramlar\u0131 \u00fczerinden analiz etmektedir. Yap\u0131sal entegrasyon, m\u00fcltecilerin okullara eri\u015fimini ve e\u011fitim hizmetlerini kullanmalar\u0131n\u0131 ifade ederken, ili\u015fkisel entegrasyon, m\u00fcltecilerin ve ulusal \u00f6\u011frencilerin sosyal olarak nas\u0131l b\u00fct\u00fcnle\u015fti\u011fini ve topluma entegre olduklar\u0131n\u0131 incelemektedir.\n\n\u00c7al\u0131\u015fma, \u00fc\u00e7 farkl\u0131 \u00fclkede uygulanan e\u011fitim modellerini detayland\u0131rmakta ve bu \u00fclkelerdeki m\u00fcltecilerin e\u011fitime eri\u015fim durumlar\u0131, e\u011fitim kalitesi ve topluma entegrasyon s\u00fcre\u00e7lerine dair bulgular sunmaktad\u0131r. Ayr\u0131ca, m\u00fclteci e\u011fitiminin kar\u015f\u0131la\u015ft\u0131\u011f\u0131 temel zorluklara ve bu zorluklarla ba\u015fa \u00e7\u0131kabilmek i\u00e7in \u00f6nerilen ba\u015far\u0131l\u0131 uygulamalar\u0131 s\u0131ralamaktad\u0131r. E\u011fitimde kalite, eri\u015fim ve aidiyet gibi kilit kavramlar etraf\u0131nda d\u00f6nen bu analiz, m\u00fcltecilere y\u00f6nelik e\u011fitim politikalar\u0131n\u0131n nas\u0131l daha etkili hale getirilebilece\u011fi konusunda \u00f6nemli \u00f6nerilerde bulunmaktad\u0131r.\n\n\u00d6zetle, bu \u00e7al\u0131\u015fma, m\u00fclteci e\u011fitimine dair kapsaml\u0131 bir de\u011ferlendirme yaparak, \u00e7e\u015fitli \u00fclkelerdeki uygulamalar\u0131 kar\u015f\u0131la\u015ft\u0131rmal\u0131 bir \u015fekilde incelemekte ve m\u00fcltecilerin e\u011fitim sistemlerine entegrasyonunu kolayla\u015ft\u0131rmak i\u00e7in gereken politik ve uygulamal\u0131 ad\u0131mlar\u0131 ortaya koymaktad\u0131r.","behavior":"degenerate_repetition","conversation_hash":"dca5a75ef8e2b4a040275983ec758535","conversation_redacted":true,"conversation_toxic":false,"country":"T\u00fcrkiye","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Turkish","model":"gpt-4-turbo-2024-04-09","n_chars":118034,"n_turns_in_conversation":3,"n_words":16170,"source_row_offset":52542,"source_shard":"turns-00024.parquet","timestamp":"Thu, 27 Jun 2024 19:37:23 GMT","turn_id":"95dd27b579871a9587cc5c34","turn_index":1,"user_text":"alttaki metni analiz edip \u00f6zetler misin? literat\u00fcr taramas\u0131 yap\u0131yorum:\n\nBy Sarah Dryden-Peterson et al. 2018\nINCLUSION OF REFUGEES\nIN NATIONAL EDUCATION SYSTEMS\nThis paper was commissioned by the Global Education Monitoring Report as background information to assist in drafting the 2019 GEM Report, Migration, displacement and education: Building bridges, not walls. It has not been edited by the team. The views and opinions expressed in this paper are those of the author(s) and should not be attributed to the Global Education Monitoring Report or to UNESCO. The papers can be cited with the following reference: \u201cPaper commissioned for the 2019 Global Education Monitoring Report, Migration, displacement and education: Building bridges, not walls\u201d. For further information, please contact <PRESIDIO_ANONYMIZED_EMAIL_ADDRESS>.\nED/GEMR/MRT/2018/P1/6\nBackground paper prepared for the 2019 Global Education Monitoring Report\nMigration, displa\ncement and education:\nBuilding bridges not walls\n2\nCONTENTS\nAcknowledgements ...................................................................................................................................... 3\nAcronym List ................................................................................................................................................. 4\nAbstract ......................................................................................................................................................... 5\n1. Introduction ............................................................................................................................................. 6\n2. Key Concepts ........................................................................................................................................ 10\n2a. Inclusion: \u201cStructural Integration\u201d and \u201cRelational Integration\u201d ................................................. 10\n2b. Access ............................................................................................................................................. 11\n2c. Quality .............................................................................................................................................. 11\n2d. Belonging and Cohesion ................................................................................................................ 13\n3. Methods ................................................................................................................................................ 15\n4. Models of Inclusion .............................................................................................................................. 17\n5. Case Studies: Uganda, Kenya, Lebanon ............................................................................................. 20\n5a. Uganda ............................................................................................................................................ 20\nContext ............................................................................................................................................... 20\nAccess ................................................................................................................................................ 21\nQuality ................................................................................................................................................ 22\nBelonging and Cohesion ................................................................................................................... 22\n5b. Kenya .............................................................................................................................................. 24\nContext ............................................................................................................................................... 24\nAccess ................................................................................................................................................ 26\nQuality ................................................................................................................................................ 27\nBelonging and Cohesion ................................................................................................................... 28\n5c. Lebanon ........................................................................................................................................... 30\nContext ............................................................................................................................................... 30\nAccess ................................................................................................................................................ 32\nQuality ................................................................................................................................................ 33\nBelonging and Cohesion ................................................................................................................... 35\n6. Challenges and promising practices for inclusion of refugees in national education systems ...... 38\n7. Synthesis: Inclusion of refugees in national education systems ....................................................... 41\nReferences ................................................................................................................................................ 44\n3\nACKNOWLEDGEMENTS\nOur thanks to <PRESIDIO_ANONYMIZED_PERSON>, Bilal Barakat, Katarzyna Kubacka, Sonia Gomez, Mary Mendenhall, and Ann Scowcroft for feedback on an earlier draft. Thank you also to Matt Miller, at the Harvard Graduate School of Education, who facilitated inclusion of this research and writing process as a learning opportunity for students. 1\n1 Written by By Sarah Dryden-Petersona, Elizabeth Adelmana, Sagra Alvaradoa, Katelin Andersona, Michelle J. Bellinob, Ranya Brooksa, Sayeda Unsa Shah Bukharia, Elizabeth Caoa, Vidur Chopraa, Zuhra Faizia, Ben Gullaa, Dahlia Maaroufa, Celia Reddicka, Ben Scherrera, Elizabeth Smoakea, and Elli Suzukia\na Harvard Graduate School of Education\nb University of Michigan School of Education\n4\nACRONYM LIST\nCRRF Comprehensive Refugee Response Framework\nDRC Democratic Republic of Congo\nGES Global Education Strategy\nMDG Millennium Development Goals\nMEHE Ministry of Education and Higher Education\nPMU Project Management Unit\nSDG Sustainable Development Goals\nUN United Nations\nUNHCR United Nations High Commissioner for Refugees\nUNICEF United Nations International Children\u2019s Emergency Fund\nUNFPA United Nations Population Fund\nUNWFP United Nations World Food Program\nUNESCO United Nations Educational, Scientific, and Cultural Organization\nUNRWA United Nations Relief and Works Agency for Palestinian Refugees\nUSAID United States Agency for International Development\n5\nABSTRACT2\nThe inclusion of refugees into national education systems has provided critical opportunities for increased access to education for refugees. The everyday experiences and processes of inclusion have also raised challenges related to the quality of education for refugees and the nationals amidst whom they learn and to relational dimensions, including related to belonging and social cohesion. Drawing on original data from a 14 country study and three in-depth country case studies, we identify three models of inclusion in refugee education: shared space, geographically separate space, and temporally separate space. We examine these models through the lenses of \u201cstructural integration\u201d \u2013 access to institutions and services \u2013 and \u201crelational integration\u201d \u2013 access to inclusive identities and connectedness. We explore possibilities for models of inclusion in refugee education to foster both structural and relational integration vis-\u00e0-vis the concepts of access, quality, and belonging and cohesion and present an agenda that can guide research, policy, and practice in this field moving forward.\n2 Written by Sarah Dryden-Petersona, Elizabeth Adelmana, Sagra Alvaradoa, Katelin Andersona, Michelle J. Bellinob, Ranya Brooksa, Sayeda Unsa Shah Bukharia, Elizabeth Caoa, Vidur Chopraa, Zuhra Faizia, Ben Gullaa, Dahlia Maaroufa, Celia Reddicka, Ben Scherrera, Elizabeth Smoakea, and Elli Suzukia\naHarvard Graduate School of Education bUniversity of Michigan School of Education\n6\n1. INTRODUCTION\nBy the end of 2016, the number of people who lived as refugees, outside their country of origin and due to a well-founded fear of persecution, totaled approximately 22.5 million (UNHCR, 2017f). Almost three and a half million of them were newly displaced in 2016 alone, primarily from conflict in Syria, but also due to conflicts in Iraq, Mali, Burundi, and South Sudan. This new wave of refugees joined millions of others who have been displaced for multiple decades, from protracted conflicts in countries such as Afghanistan, Democratic Republic of Congo (DRC), and Somalia. More than half of refugees globally are children, under the age of 18 (UNHCR, 2017f).\nThe right to education for all children, including refugees, is articulated in international instruments, such as the Universal Declaration of Human Rights and the Convention on the Rights of the Child. The United Nations High Commissioner for Refugees (UNHCR) is the global organization mandated with protecting refugees\u2019 rights, and with delivering assistance such as food, shelter, and water, and enabling necessary services, including education.3 UNHCR\u2019s work on education, as on other areas of relief, is coordinated with host governments, the nation-states in which refugees reside. The 1951 Convention relating to the Status of Refugees specifically asserts refugees\u2019 right to education, but within a framework of existing provision in the host country: signatory states \u201cshall accord to refugees the same treatment as is accorded to nationals with respect to elementary education\u2026 [and] treatment as favourable as possible\u2026 with respect to education other than elementary education\u201d (UNHCR, 2011). The 2016 New York Declaration for Refugees and Migrants reasserts this right to education, for both primary and secondary schooling (United Nations General Assembly, 2016, p. 14).\nEnsuring the right to education for refugees is an urgent challenge for the global community. Refugees access education at lower rates than other children globally: in 2016, only 61 percent of refugee children accessed primary school, compared to 91 percent of all children globally; at the secondary level, 23 percent of refugee children accessed education, whereas 84 percent of young people did globally (UNHCR, 2017g). Policy actors working toward achieving equity in learning, and the targets established within the Sustainable Development Goals frameworks, recognize the many ways in which the inconsistent\n3 We do not include Palestinians in this analysis of inclusion of refugees in national education systems, as education of Palestinians is led by the United Nations Relief and Works Agency (UNRWA) in separate education systems.\n7\nrealization of the right to education for refugees continues to exacerbate inequities in education access and outcomes, eventually undermining the very essence of these goals. This recognition has been central to recent policy shifts in refugee education. At the same time, current focus on refugee education also importantly illuminates remaining and ongoing challenges in meeting the right to education for the most marginalized national children with whom refugees usually live and go to school.\nPrior to 2012, in most settings, refugees were educated in parallel schools, separate from national students and often following the curriculum and in the language of instruction of the country of origin (Dryden-Peterson, 2016). More recently, the UNHCR Global Education Strategy (GES) 2012-2016 articulated a new approach to the education of refugees: inclusion in national education systems (UNHCR, 2012, p. 8). Inclusion, or the process of coming together through refugee learners\u2019 access to government schools and/or the curriculum followed by the governments of the host countries, intends to increase refugee learners\u2019 access to education. In doing so, it simultaneously addresses several key dimensions of contemporary displacement. First, refugee children are likely to spend their entire schooling years in a host country, given the protracted nature of displacement; the average length of exile for refugees is estimated between ten and twenty-five years, and to be up to three times as long as it was in the early 1990s (Crawford, Cosgrave, Haysom, & Walicki, 2015; Devictor & Do, 2016; Milner & Loescher, 2011). Second, refugee-only schools are impractical given the increasingly urban residence of refugees; by the end of 2015, over 60 percent of refugees lived in urban areas (UNHCR, 2016, p. 53). Third, refugee-only schools are unsustainable over the longer-term given persistent shortfalls and unpredictability in funding for refugee education; less than two percent of all humanitarian funding is allocated to education (Global Education Monitoring Report, 2017a, pp. 7-8).\nIn our research, we have documented a swift adoption of this approach of inclusion of refugees to national education systems (Dryden-Peterson, 2015, 2016; Dryden-Peterson, Adelman, Bellino, & Chopra, Under review). Some nation-states had already moved toward including refugees within their systems of education prior to the introduction of the GES. For example, the Kenyan curriculum was used in refugee schools in Dadaab camps in Kenya as early as 1997 (Dryden-Peterson, Dahya, & Adelman, 2017), and national schools in urban areas of Uganda admitted refugees in ad hoc ways as early as the 1990s (Dryden-\n8\nPeterson & Hovil, 2004).4 In other nation-states, the shift to inclusion was rapid in response to the GES. In 2010, only 5 of 14 of the largest refugee-hosting nation-states5 used the national curriculum and national languages of instruction to teach refugee learners, but by 2014, 11 of these 14 nation-states did (Dryden-Peterson, 2016). Relationships between national governments, education authorities, and UNHCR also developed over this time. Prior to the introduction of the GES, UNHCR did not have a relationship with a single government authority in the education sector in any country in which it worked (Dryden-Peterson, 2011b). By 2016, UNHCR had formal relationships with national authorities in 20 of their 25 priority country operations.6 Because of these relationships, UNHCR was able to negotiate inclusion of refugees in existing systems of education through either access for refugees to government schools and/or curriculum, and established means of coordination with governments (Dryden-Peterson, 2016). The December 2017 Djibouti Declaration \u2013 signed by Ministers of Djibouti, Ethiopia, Kenya, Somalia, South Sudan, Sudan, and Uganda \u2013 further sets the goal to \u201cintegrate education for refugees and returnees into National Education Sector Plans by 2020\u201d (IGAD Member States, 2017, p. 3).\nSince the adoption of this approach, the inclusion of refugees into national education systems has provided critical opportunities for increased access to education for refugees. The everyday experiences and processes of inclusion have also raised challenges related to the quality of education broadly and to relational dimensions, including how refugees and nationals develop their own individual senses of belonging and well-being as well as broader social cohesion. Importantly, the vast majority of refugees \u2013 84 percent \u2013 live in countries that neighbor their conflict-affected countries of origin (UNHCR, 2017f). These host countries are generally characterized by over-stretched education systems and often fragile political and economic institutions (Dryden-Peterson, 2015; Hathaway, 2016). In these contexts, national\n4 We use the term camp to describe geographic spaces in Kenya where refugees live and access services; we use the term settlement to describe these spaces in Uganda and informal settlements to describe these spaces in Lebanon. These terms align with the terms in use in these nation-states.\n5 These 14 countries included Bangladesh, Chad, Egypt, Ethiopia, Iran, Kenya, Lebanon, Malaysia, Pakistan, Rwanda, South Sudan, Sudan, Uganda, and Yemen.\n6 These 25 priority countries were countries implementing the 2012-2016 Education Strategy. They included: Bangladesh, Burkina Faso, Burundi, Chad, Democratic Republic of Congo, Djibouti, Egypt, Ethiopia, Iraq, Iran, Jordan, Kenya, Lebanon, Malaysia, Niger, Pakistan, Rwanda, South Sudan, Sudan, Syria, Tanzania, Turkey, Uganda, Yemen, and Zambia (UNHCR, 2015, p. 9).\n9\neducation systems struggle to meet goals of quality for national students, let alone new refugee students. Sharing scarce resources within already fragile institutional settings often leads to tensions among individuals and groups (e.g., Cederman, Weidmann, & Gleditsch, 2011; Novelli, 2016; Stewart, Brown, & Langer, 2008).\nLess well-understood are the ways in which the approach of inclusion of refugees in national education systems is enacted at national levels and experienced by teachers and students in schools. These processes are the subject of this Background Paper. Below, we outline Key Concepts that guide our analysis: \u201cstructural integration\u201d and \u201crelational integration\u201d as connected to inclusion; access; quality; and belonging and cohesion. Following our methodology, we present a typology of Inclusion Models based on our empirical research, identifying three models of inclusion: shared space, geographically separate space, and temporally separate space. In the remainder of the paper, we examine these models in three nation-state case studies focusing on Uganda, Kenya, and Lebanon, and outline the challenges and promising practices that emerge across the cases. We conclude by examining the possibilities for inclusion that fosters both structural and relational integration vis-\u00e0-vis the concepts of access, quality, and belonging and cohesion and present an agenda that can guide research, policy, and practice in this field moving forward.\n10\n2. KEY CONCEPTS\n2a. Inclusion: \u201cStructural Integration\u201d and \u201cRelational Integration\u201d\nWe define inclusion, in its broadest sense, as the process of coming together. Our analysis is based in a distinction that has emerged in literature on processes of inclusion, between \u201cstructural integration\u201d and \u201crelational integration\u201d (Korac, 2003; Strang & Ager, 2010). We have chosen to use the term \u201cinclusion\u201d in this Background Paper when referring to the policies and practices that aim for refugees to have access to national systems of education, while drawing on conceptual ideas of \u201cintegration,\u201d both structural and relational. In the 2012-2016 GES, UNHCR used the term \u201cintegration\u201d, yet has shifted to using the term \u201cinclusion\u201d in its education work; the term \u201cinclusion\u201d is what is used in the Comprehensive Refugee Response Framework (CRRF) as well.7 We recognize the important political implications for host countries, by which the term \u201cintegration\u201d can imply long-term legal status or citizenship, which is not intended by a policy that opens national schools to refugee students. In this Background Paper, we use the term \u201cinclusion\u201d to describe the active and dynamic processes of coming together of refugees and nationals in schools, but draw on analytic concepts of \u201cstructural integration\u201d and \u201crelational integration\u201d to enable a focus on the educational goals of inclusion, including access, quality, and belonging and social cohesion.\n\u201cStructural integration,\u201d often also called functional integration, is defined by the ability to access institutions and services, such as education. \u201cRelational integration\u201d is a sociocultural process, related to identity development and transformation; it includes both an individual-level sense of belonging, or connectedness, as well as group-level social cohesion (e.g., Korac, 2003; Strang & Ager, 2010). The relational elements of inclusion have been largely ignored in the development and initial implementation of policies and practices that structurally include refugees in national education systems (Dryden-Peterson & Hovil, 2004; Rowley, Burnham, & Drabe, 2006). Focus has remained on the structures of inclusive services \u2013accessing national exams and certification, using the national curriculum and languages, and being taught by national teachers. While these structural elements of inclusion are foundational, we also focus in this paper on the relational elements of inclusion. We further elaborate on these ideas of\n7 Other terms in use for policies and practices of enabling access to national schools for refugee students include \u201cmainstreaming\u201d and \u201charmonization.\u201d\n11\n\u201cstructural integration\u201d and \u201crelational integration\u201d in connection to inclusion in each of the case studies and in Section 7 of this paper.\n2b. Access\nWe define access as the ability to enroll in formal, accredited education. A global commitment to universal access to education began in 1990 with the Education for All Declaration and was incorporated centrally in the Millennium Development Goals (MDGs). Since 1990, the number of children without access to school has dropped from 108 to 51 million (UNESCO, 2014, p. 5). Barriers to accessing education are of two general types: physical and identity-based. Physical barriers encompass resources and capacity, such as limitations in the number of school buildings or teachers, and include the unequal distribution of these resources geographically. Identity-based barriers encompass differential access to school and discrimination as a result of ethnicity, gender, language, religion, sexuality, or status as a non-citizen or refugee (e.g., Lewin, 2009; Lloyd, Mete, & Sathar, 2005; Save the Children, 2016). These physical and identity barriers often intersect, making some students multiply-marginalized.\nInclusion of refugees in national education systems could address some of these physical and identity-based access barriers, thereby increasing access to formal schooling for refugees. For example, within a pre-existing education system, refugees could have greater access to education if the infrastructure of school buildings and a teaching corps were already in place. On the other hand, access to education for refugees could be limited in situations where national education systems are under-resourced, with already overcrowded classrooms and limited numbers of teachers.\n2c. Quality\nQuality education contributes to children\u2019s whole development, allowing them to learn well, develop, and thrive. Quality is a central goal of education development broadly as articulated in the Sustainable Development Goals (SDGs) and the Incheon Declaration and Framework for Action (UNESCO, 2016; United Nations, 2016). It was also clearly articulated in the Comprehensive Refugee Response Framework, wherein the United Nations General Assembly reaffirmed the commitment of member states to provide\n12\nquality primary and secondary education in safe learning environments for all refugee children (United Nations General Assembly, 2016, p. 14).\nIn an extensive review of the literature on educational initiatives to improve learning outcomes in developing countries, Ganimian and Murnane (2016) concluded that the general failure to improve education quality may stem from the approach of offering \u201cmore of the same\u201d to vastly expanded student populations (see also, Nicolai, Wild, Wales, Hine, & Engel, 2014, for a discussion of the incentives for quality initiatives). We argue that these challenges to quality are also salient in the case of refugee education, where student populations are not only vastly expanded within a relatively short period of time, but also heterogeneous along dimensions of, for example, educational background, experiences of trauma, and language. We focus on the role of teachers and school administration in our analysis of the quality of education, as critical mechanisms for shifting from a \u201cmore of the same\u201d approach to \u201cchang[ing] children\u2019s daily experiences\u201d in schools (Ganimian & Murnane, 2016, p. 744) in ways that hold promise for increased learning.\nA national education system could create conditions for quality education for all learners, including refugees, through practices related to curriculum, pedagogy, and language, all highly dependent on teachers and school administration. Existing curriculum and trained teachers to implement it could assist refugee young people in acquiring the skills and knowledge they need to pursue further education and/or economic livelihoods (Schweisfurth, 2015). In situations where curriculum poses challenges of relatability for students, skilled teachers, with pedagogical training provided within a national system, can adapt curriculum to enable relevance (Bajaj & Bartlett, 2017; Bellino & Williams, 2017; Dryden-Peterson & Sieb\u00f6rger, 2006; Paulson, 2015). As related to language, teachers with training in language-learning and multilingual teaching are better able to meet the diverse linguistic needs of students, refugee and national alike (Bartlett & Garcia, 2011; Ruiz, 1984). On the other hand, we also recognize that if the quality of education within a national system is low, as related to the content of the curriculum, the pedagogical skills of teachers, and the school\u2019s capacity to address the multiplicity of linguistic and ethnic diversities, inclusion may not result in quality education for refugee learners.\nData on learning outcomes for refugees \u2013 as for nationals \u2013 is often limited, with particular challenges in settings of inclusion where national data is not disaggregated by refugee status. Therefore, we include reference to learning outcomes only in a few instances.\n13\n2d. Belonging and Cohesion\nSocial cohesion is a central mechanism in the achievement of \u201cinclusive communities\u201d envisioned by the SDGs (United Nations, 2016). Inclusive and cohesive communities work toward promoting the well-being of all members, tackle exclusion and marginalization, foster belonging, promote trust and offer opportunities for upward, social mobility (OECD, 2011). Fostering cohesion is fundamental to the policies that have prompted inclusion of refugees into government schools, echoed in the Comprehensive Refugee Response Framework. This framework advocates that \u201c[d]iversity enriches every society and contributes to social cohesion,\u201d while countering \u201cracism, racial discrimination, xenophobia and related intolerance against refugees and migrants.\u201d Central to the ways in which diversity can play this role is \u201cdirect personal contact between host communities and refugees and migrants\u201d (United Nations General Assembly, 2016, p. 3). Important to note is that schools alone cannot create \u201cinclusive communities,\u201d as they operate within larger social and political structures that are enabling or constraining in the messages they send to refugee children and their families about the possibilities of belonging (Antonsich, 2010; Banks, Forthcoming; Bellino & Dryden-Peterson, Under review; Janmyr, 2016). Greater engagement of refugee communities in policymaking and planning as connected to models of inclusion could ensure that these dimensions of the broader environment related to belonging and cohesion are addressed (see, for example, Barbelet, Hagen-Zanker, & Mansour-Ille, 2018; Plan International, 2017).\nThough belonging is a multifaceted construct, we conceptualize it here to refer to individuals\u2019 sense of safety, well-being, and membership as cultivated through productive relationships (Antonsich, 2010; Yuval-Davis, 2006). Inclusion of refugees and nationals together in schools could be a critical space for direct contact between individuals, creating possibilities for refugees and nationals to cultivate an individual sense of belonging, shared membership in a learning community, and to negotiate their relationships with each other. Contact alone, however, does not build broader social cohesion (Allport, 1979 [1954]). School structures, practices, and relationships contain explicit and implicit messaging of norms of who belongs and who does not (e.g., Abu El-Haj, 2015; Banks, Forthcoming; Bellino & Williams, 2017; Reijerse, Van Acker, Vanbeselaere, Phalet, & Duriez, 2013). Inclusion of refugees in national schools could support or undermine social cohesion, depending on the curriculum, pedagogy, and, critically, the relationships built therein (e.g., Bajaj & Bartlett, 2017; Bar-Tal & Rosen, 2009; Bellino, 2017).\n14\nLanguage likely plays a central role in refugee children\u2019s ability to construct a sense of belonging when included in national schools. In examining the links between language and belonging and cohesion, we deliberately adopt a broad understanding of language. We refer to language not only as shared, verbal scripts that facilitate inter-personal communication and understanding, but also the non-verbal, socio-cultural scripts and practices that enable refugee learners to understand the \u201crules of the game\u201d in their host countries, the unspoken and often invisible social rules and norms that are not explicitly taught. A combination of both scripts is essential for refugees\u2019 \u201crelational integration.\u201d In multilingual host countries, the many languages present in the communities and schools into which refugees are included present both an opportunity and a challenge to refugee students\u2019 access to quality education, mediated by connections between language and identity (Benson, 2012; Garcia, 2012; Shin, 2013). Early instruction in the home language can support children\u2019s development of a positive self-image as well as their academic learning, preparing them to acquire foreign languages in the later years of schooling (Cummins, 1981, 2000). Refugee students\u2019 sense of belonging when included in national schools may be closely connected to their ability to both maintain languages used in the country of origin and learn languages used in the host country, allowing them to communicate with national teachers and peers and successfully navigate host country structures \u2014skills that are critical for facilitating learning and belonging \u2014 while also maintaining relationships with family.\nIn the empirical case studies presented in this Background Paper, we focus on evidence of ways in which schools enable students to cultivate individual senses of belonging, and we return in our synthesis to the implications for broader social cohesion.\n15\n3. METHODS\nOur study is comparative and draws on case methods (George & Bennett, 2005; Gerring, 2004). We analyzed an original dataset on inclusion of refugees in national education systems, drawing on several larger research projects on this topic, with data collected between 2012 and 2017 (Adelman & Dryden-Peterson, Under review; Bellino & Dryden-Peterson, Under review; Chopra, 2018; Chopra & Adelman, 2017; Dryden-Peterson et al., Under review). We draw on analysis from these larger projects to frame approaches to refugee education globally. For this Background Paper, we also conducted new desk-based analysis of field-based data and existing literature, with a focus on three nation-states in which refugees are included in national education systems in various ways: Uganda, Kenya, and Lebanon. The dataset for this Background Paper comprised publicly available documents related to policy, strategy, and legal frameworks, at global, national, and local levels; original interviews with key stakeholders, governments, United Nations agencies, and Non-Governmental Organizations; original interviews with teachers of refugees; and original classroom observations in refugee-serving schools.\nA team worked collaboratively to conduct analysis and writing for this paper, under the leadership of Sarah Dryden-Peterson. The team included Sagra Alvarado, Katelin Anderson, Ranya Brooks, Sayeda Unsa Shah Bukhari, Elizabeth Cao, Ben Gulla, Dahlia Maarouf, Ben Scherrer, Elizabeth Smoake, and Elli Suzuki, graduate students enrolled in a for-credit course at the Harvard Graduate School of Education; and Elizabeth Adelman, Michelle Bellino, Vidur Chopra, Zuhra Faizi, and Celia Reddick, who were involved in field-based data collection and writing.\nWe developed a coding system that was informed by our literature review and the Terms of Reference for the paper, as well as by the purposes of Background Papers as both technical and advocacy tools, with the aim that our work be actionable in the context of the Report and the work of policymakers and practitioners in refugee education. We identified six central codes: model of inclusion, access, quality, belonging, educational outcomes, and policy. Each of these codes were further refined such that quality, for example, included sub-codes of curriculum, language, pedagogy, safety, and teachers. We paid particular attention to the experiences of individuals, in line with a cross-cutting issue of the 2019 Global Education Monitoring (GEM) Report, \u201cIn what ways can the voices of migrants improve our understanding of migration and education?\u201d (Global Education Monitoring Report, 2017b, p. 5).\n16\nWe used the Dedoose platform to collaboratively code our data, working in groups, organized by country, of three to four researchers and using a model of \u201ccollaborative coding,\u201d where researchers reviewed each other\u2019s coding to discuss and reach a consensus on the code applied (Smagorinsky, 2008). After coding all data, we wrote analytic memos on each sub-code, using matrices to help further organize the data and identify connections between ideas within each case and across the three country cases.\n17\n4. MODELS OF INCLUSION\nCountries have adapted the GES policy of inclusion in various ways. Our study of 14 countries finds that the resulting models of refugee education range from full access to government schools with refugees and nationals together in the same classrooms, to access to government schools but separation from nationals either geographically or temporally, to no access to government schools (i.e., no inclusion). We found that the model adopted in each host country reflected the policy environment in the host country as well as the conflict dynamics in and between the countries of origin and host countries, particularly the perceived duration of the conflict refugees had fled and historical and contemporary relationships between refugees and the host country.\nAcross these 14 countries, we identified dimensions of \u201cstructural integration\u201d such as use of national curriculum; instruction by national teachers; use of national language(s) of instruction; and access to national certification (Dryden-Peterson et al., Under review). We identified three models of \u201cstructural integration,\u201d that enable refugee learners\u2019 access to the structures of the government school system present in their host countries. These models are: (1) shared space, in which refugees and nationals are physically together in the same schools and classrooms; and (2) separate space, which for analytic purposes we divide into two sub-models (2a) geographically separate, in which refugees and nationals reside in different geographical areas and thus attend different schools; and (2b) temporally separate, in which refugees and nationals attend the same schools but at different times, often referred to as a \u201cshift system.\u201d Table 1 describes the dimensions of \u201cstructural integration\u201d, and in Table 2 we present the models of inclusion for our cases of Uganda, Kenya, and Lebanon. While we draw clear definitions of these models, in practice, the characteristics of these models are more blurred and nuanced, which we explore in the case studies that follow. We also explore possibilities for \u201crelational integration\u201d in conjunction with each models of inclusion.\n18\nTable 1. Elements of \u201cStructural Integration\u201d Present in Policy National curriculum National teachersa National language of instruction National certification\nUganda\n\u2713\n\u2713+ refugee \u201cincentive\u201d teachersb\n\u2713\n\u2713\nKenya\n\u2713\n\u2713 + refugee \u201cincentive\u201d teachers\n\u2713\n\u2713\nLebanon\n\u2713\n\u2713\n\u2713\n\u2713\na The individuals who teach refugees, while nationals, often do not have the civil service status of teacher vis-\u00e0-vis being on long-term government salary (rather than on a short-term contract) and are therefore not included within multi-year education budgets.\nb Refugee \u201cincentive\u201d teachers are refugees who, without the right to work, are compensated with an incentive wage rather than a salary. In none of the case study countries are refugee teachers able to work as government-salaried teachers.\n19\nTable 2. Models of Inclusion, per Policy, and Possibilities for \u201cRelational Integration\u201d Shared Space Separate Space Geographically separate Temporally separate\nUganda\nRefugees and nationals have access to government schools regardless of legal status. Refugees and nationals attend school together in urban areas and at times in the periphery of settlements.\nMost refugees live in rural settlements, where few nationals reside, and attend government schools constructed in the settlements or nearby government schools.\nKenya\nRefugees and nationals have access to government schools. Refugees and nationals attend school together at times in the periphery of camps and predominantly in and around Nairobi.\nMost refugees live in camp settings and attend camp-based schools, which are government schools but established primarily for refugees and intended to include 10 percent national students.\nLebanon\nLebanese national and Syrian refugee students attend the same schools, but at different times of the day. Lebanese national students attend school in the morning (first) shift, and nearly all Syrian refugee students attend school in the afternoon (second) shift.a\na According to the Lebanon MEHE, as of November 2017, there were 31,930 non-Lebanese students enrolled in the first shift. Data is not available on the nationality or refugee status of these non-Lebanese students, nor on whether there are Lebanese nationals in the first shifts at the schools that enroll non-Lebanese students in the first shift or if they are exclusively non-Lebanese (Lebanon Ministry of Education and Higher Education, 2017a). By policy, Syrian refugee students attend the second shift (Lebanon Ministry of Education and Higher Education, 2016).\n20\n5. CASE STUDIES: UGANDA, KENYA, LEBANON\n5a. Uganda\nCONTEXT\nUganda is surrounded by nation-states where national borders, like Uganda\u2019s, were drawn during colonialism, and that have subsequently experienced some of the most intense and protracted conflicts in recent history (Albaugh, 2014). As a result of persistent fragility and conflict in the region, since 1955 Uganda has hosted large numbers of refugees from various countries in central and eastern Africa. As of this writing, Uganda hosts the largest number of refugees in Africa, with 1.38 million refugees living in the country, primarily from South Sudan, Burundi, and the Democratic Republic of the Congo (DRC) (UNHCR, 2017d). Refugees arriving from South Sudan are comprised of primarily women and children fleeing ongoing armed conflict as well as food insecurity; the vast majority of the current South Sudanese refugee population in Uganda have arrived since 2014 (United Nations Security Council, 2017a). Burundian refugees continue to flee Burundi for Uganda due to ongoing political violence and human rights violations, including extrajudicial killings, torture, and sexual violence (United Nations General Assembly, 2017), and refugees from DRC have left their country of origin because of decades-long political and ethnic violence as well as human rights abuses perpetrated by both the government and armed groups (United Nations Security Council, 2017b).\nUganda became a signatory to the 1951 Refugee Convention in 1987. However, until recently, the 1960 Uganda Control of Alien Refugees Act remained in force, limiting the right to freedom of movement and the right to own property for refugees (Mulumba, 2018). With the passage of the 2006 Refugee Act and the subsequent 2010 Refugee Regulations, refugees have access to a greater range of rights and services in Uganda, and Uganda has been cited as an exemplar model for inclusive refugee policy in the region (Betts & Collier, 2017; World Bank Group, 2016). Provisions for refugees include the right to work and access to basic services like primary education and healthcare (World Bank Group, 2016). In Uganda, refugees reside both in urban areas and in a large number of settlements located primarily in the north and west of the country.\n21\nIn 2007, Uganda adopted a thematic curriculum for the first three years of primary school, through which academic content is organized around themes that are familiar to learners. The curriculum calls for home language instruction in the early years of schooling. Students also learn English as a subject, and switch to English as the language of instruction in the later years of primary school (National Curriculum Development Centre, 2018). Despite these curriculum reform efforts, challenges to educational access and quality remain. In 2016, the World Bank reported that 78 percent of primary school children drop out of school nationally and 40 percent of secondary school children do so (World Bank, 2017). Data regarding learning outcomes in Uganda has not been disaggregated between national and refugee students.\nUnder the Refugee Act of 2006, refugee students have access to government schools (\"The Refugees Act 2006,\" 2006). By policy, refugees are to be treated no differently than nationals, accessing shared-space schools in which national and refugee students learn together in the same classrooms. However, in practice, most refugee students attend geographically separate schools given their residence in settlements where few nationals reside. Refugees in Uganda have varied but low access to education pre-migration, with reported primary enrollment rates of 50 percent in South Sudan (UNICEF, 2017), 49 percent in Burundi (World Bank, 2017), and a lack of data on enrollment in the conflict-affected regions of DRC from which refugees have fled.\nACCESS\nDespite being legally permitted to access Ugandan schools, refugees enroll at lower rates than do national students. While data are limited and rates vary across regions of Uganda, more than 90 percent of nationals access primary school, only about one third of refugees do; at secondary levels, 20 percent of school-aged nationals are enrolled in school while 16 percent of refugee children are (Dryden-Peterson, 2015; UNESCO Institute of Statistics, 2018). Among the barriers to accessing education for refugees are limited resources and capacity. Schools in or near refugee settlements, which serve mainly refugees, are often understaffed, leading to overcrowding, with an average student-teacher ratio of 150:1 (Government of Uganda, United Nations, & UNHCR, 2017). At the secondary level these challenges are further exacerbated for both national and refugee students. Shortages of secondary school teachers and staff are common across most districts (Government of Uganda et al., 2017), prompting school officials to hire untrained and underqualified teachers to enable access for more children, national and refugee alike, even if this negatively affects quality. Furthermore, existing schools have limited availability of classrooms,\n22\ndesks, and school materials. Inadequate funding compounds the problem of access, as schools often lack the capital to pay teachers and provide them with training (Government of Uganda et al., 2017).\nQUALITY\nIn 2010, 12.7 percent of primary school teachers in Ugandan government schools were uncertified (UNESCO, 2017). There is little research that documents whether Ugandan national teachers are specifically trained to work with refugee students in government schools, although NGOs in Uganda regularly provide training to national teachers during emergencies, focusing on early childhood development and accelerated learning programs (Government of Uganda et al., 2017; Nicholson, 2006).\nWithin overcrowded classrooms, student-centered pedagogical methods are rare (Trudell, 2016). There are consequently few opportunities for students, refugee or national, to get individualized attention or to regularly participate in classroom discussions (Dryden-Peterson, 2017). Multilingual language of instruction policies also present a challenge for education quality, with 68 different languages spoken in homes and communities and 21 of these used as languages of instruction in public schools in the early years of primary school (Altinyelken, Moorcroft, & van Der Draai, 2014; UNICEF, 2016; Ward, Penny, & Read, 2006). In practice, Ugandan national teachers often struggle to implement home language instruction because of a lack of sufficient training or materials (Piper & Miksic, 2011; Tembe & Norton, 2008). For refugee students included in Ugandan government schools, home language instructional policies are also likely implemented with variable consistency (UNHCR, 2013). As one refugee student from DRC explained, there are not enough \u201cteachers who are capable of expressing themselves in our languages, French, and Kiswahili\u201d (Dryden-Peterson, 2006). The frequent use of English rather than home language in schools likely affects the quality of the schooling refugees receive, with negative outcomes for learning and persistence (Benson, 2012; UNHCR, 2013).\nBELONGING AND COHESION\nWhile Uganda has made strides toward \u201cstructural integration,\u201d obstacles to \u201crelational integration\u201d remain entrenched. Uganda\u2019s shared space model of inclusion enables \u201cstructural integration\u201d for refugees living in both rural settlements and urban settings. However, refugee families who have been living in Uganda still face xenophobic discrimination and physical, psychological, or sexual abuse even\n23\ndecades into their exile (Dryden-Peterson, 2011a; Dryden-Peterson & Hovil, 2004), presenting challenges for \u201crelational integration.\u201d\nIn order to promote belonging and social cohesion, Uganda has called for increased funding to provide schools and communities with counselors or staff capable of providing more directed psychological and emotional support to students (Government of Uganda et al., 2017), but this kind of support and/or additional training for teachers on these topics is not yet consistently available. We explore this issue further in Section 6 below. Additionally, some initiatives within host communities have created opportunities for national and refugee students to interact in less-structured settings, such as sports tournaments, enabling individuals and communities to build social connections (Reddick & Dryden-Peterson, 2018).\nEven when refugees in Uganda complete secondary education, opportunities for continuing their education or developing economic livelihoods can be limited. Few refugees in Uganda are enrolled in tertiary education (Government of Uganda et al., 2017). The cost of tuition and recognition of credentials from countries of origin prevent youth from accessing tertiary education and creating a pathway to economic self-reliance. While public universities charge refugees and nationals similar tuition fees, some private universities charge refugees international tuition fees. Refugees in Uganda legally have the right to work; however, without access to quality education, without language fluency, and with on-going discrimination, barriers to enacting that right remain.\nBox 1. South Sudanese Refugee in Uganda (using data derived from Reddick & Dryden-Peterson, 2018)\nAs a refugee from South Sudan, Edward entered primary school in Uganda in 2001 with no knowledge of English. His teachers were primarily Arabic-speaking refugees who had trained or taught in what was then Sudan, whereas their teaching manuals were in English. Edward recalls, \u201cMost of our former teachers could not speaking [sic] very well, they were just writing things on the blackboard, that gave us hard time.\u201d He continues, \u201cAll we did was copying, copying, copying, I was frustrated. I had to leave. That is why even up to university I had problems with my grammar, the tenses.\u201d\n24\nEdward spent his upper primary and secondary school in shared space schools in Arua, in northern Uganda, with majority national students where language continued to pose a major obstacle to learning and belonging. Edward explains that because he and his refugee peers predominantly spoke Arabic and little English \u201c[i]t was very hard for us to interact with the Ugandans in the school that we were in.\u201d Despite feeling isolated, in secondary school Edward made a Ugandan friend who shared an interest in art. Through their interactions, Edward began to learn Lugbara, his friend\u2019s home language and one of 68 languages spoken in Uganda. In university, where most of the Ugandan students spoke Luganda, Edward was faced with learning yet another language. \u201cIt was very hard for me to, like, integrate.\u201d He started to learn basic greetings in Luganda, though he still doesn\u2019t understand the language. For Edward, learning languages of the host country is essential for belonging and social cohesion. He explains that in university, \u201cI first saw the need for these local languages to be learnt by refugee students\u201d so that nationals and refugees could communicate through a common language.\n5b. Kenya\nCONTEXT\nIn Kenya, refugee services, including education, are guided by policies of encampment that date back to the early 1990s. These policies were reinforced in the Refugee Act of 2006 and the government\u2019s directive, albeit unevenly enforced, in 2012 to relocate refugees from urban areas to camps. As a result, 86 percent of the 492,761 total refugees in Kenya are located in camps, with a much smaller percentage living in urban areas (UNHCR, 2017b).8\nThere are two main camps in Kenya: Dadaab camp and Kakuma camp, each with populations of over 100,000 refugees. Dadaab refugee complex was established in 1991 when Somali refugees arrived in Kenya fleeing the civil war. Somali refugees continued to arrive in Dadaab over the past several decades,\n8 As of February 2017, UNHCR reported that 170,833 refugees reside in Dadaab camp, 164,571 refugees reside in Kakuma camp, 90,090 in Alinjugur camp, with 67,267 refugees continuing to reside in urban areas. More recently, the government announced the development of a new refugee settlement near Kakuma called Kalobeyei, in collaboration with Turkana County, hosting a population of 34,494 refugees.\n25\nfleeing drought, famine, and ongoing conflict in Somalia. Kakuma camp was established in 1992 when Sudanese refugees fled the civil war, and in the same year, Ethiopian refugees fled their country following the collapse of the Ethiopian government. In December 2013, a renewed conflict broke out in South Sudan, increasing the rate of new refugees entering Kenya. The new influx of refugees over the last decade has triggered rising public fears about national security, government directives for urban refugees to return to camps, plans for camp closures, and increased harassment of refugees by state security actors (Burns, 2010; Campbell, 2006; Government of Kenya, 2014; Pavanello, Elhawary, & Pantuliano, 2010; Wilson, 2014). Under the 1951 Refugee Convention, to which Kenya is a signatory, refugees have the rights to mobility and work in Kenya, but neither is upheld in practice (Bellino & Dryden-Peterson, Under review). Given this strict encampment policy, educational inclusion takes place in a geographically separate setting.\nRefugee students follow the Kenyan curriculum, learn in the Kenyan languages of instruction, and have access to the same school leaving certification as Kenyan nationals (see Table 1). The key stakeholders providing refugee education in Kenya are the Ministry of Education, UNHCR, UNICEF, and implementing partners including Lutheran World Federation (LWF), Don Bosco, Xavier Project (XP), Windle International Kenya (WIK), Islamic Relief Worldwide (IRW) and Care International. The majority of the stakeholders working in the education sector for refugees identify UNHCR\u2019s Global Education Strategy as a critical coordination tool. As of July 2014, all refugees in Kenya have the right to access government schools, with all camp schools falling under the jurisdiction of the local Ministry of Education district sub-county officers (Bellino & Dryden-Peterson, 2014). In practice, most refugee students attend camp schools given their residence in refugee camps. A few academically exceptional students manage to access government schools outside the camps, sponsored by UNHCR and partners.\nWhile we draw on data from Dadaab camps and Nairobi where possible, most of our analysis derives from data collection in Kakuma camps. Kalobeyei, near Kakuma, has been envisioned as a settlement where refugees and host community members will be socially and economically integrated. Currently, however, plans for the process of inclusion remain uncertain, given threats of camp closures and conflict and repatriation dynamics, as thousands of refugees have relocated from Dadaab to Kakuma, settling in Kalobeyei.\n26\nACCESS\nIn urban areas of Kenya, refugees attend government schools with nationals under the shared space model. The barriers to accessing education in Kenyan government schools for refugees often mirror those faced by marginalized nationals (Bellino & Dryden-Peterson, Under review). Barriers to access are related to cost, resource, and capacity constraints. Refugee families often do not have the economic resources to pay hidden fees associated with government schools, such as for uniforms, textbooks, meals, and extra tuition charges. Fees for secondary school substantially limit access for both nationals and refugees, especially to high quality schools (Bellino & Dryden-Peterson, 2014; Mendenhall et al., 2015).\nIn camps, policy also specifies the shared space model, reserving 10 percent of space for Kenyan nationals, in an effort to include students from the host community. In practice, however, camp schools almost exclusively cater to refugee populations, with host community students instead enrolling in nearby government schools. This results in an inclusion model that is predominantly geographically separate (see Table 2). Schools serving refugees in Kenya are severely constrained in their physical and human resources. In camp settings, classrooms are overcrowded, with often little space for teachers or students to move and challenges in hearing instruction (Dryden-Peterson et al., Under review). Teachers of refugees also serve multiple duties beyond the classroom and have inconsistent attendance (Bellino & Dryden-Peterson, Under review). Consequently, net enrollment in basic education within camps is low relative to national enrollment rates. Primary school net enrollment rate for refugees across camp settings in Kenya is roughly 70 percent for refugees, compared to 85 percent for Kenyan nationals, while secondary school net enrollment rate is 6.1 percent for refugees compared to 47.8 percent for Kenyan nationals (UNHCR, 2017e).\nAccording to national policy, refugee students in Kenya are to be taught by Kenyan teachers. However, in camp settings refugee teachers are often employed as \u201cincentive teachers,\u201d compensated through a lower wage or stipend rather than a salary, as per Kenyan labor laws, and with less job security than a government-salaried teacher. In Kakuma, at the primary level, there are 100 national teachers and 648 refugee teachers (UNHCR, 2017c). Employing a mixed teaching staff compensates for shortages of national teachers in the region, though introduces challenges to quality, further described in the next section.\n27\nQUALITY\nOne dimension of student learning is student achievement tests. A learning outcome that is particularly important in the context of Kenya is the primary school leaving exam, the Kenya Certificate of Primary Education (KCPE), which all Kenyan nationals and refugee students take at the end of primary education. Recent results from the KCPE in Kakuma are promising. In 2016, 90 percent of both girls and boys sitting for the KCPE passed with scores enabling their transition to secondary school, far greater than the 76 percent national pass rate (UNHCR, 2017a). Further research is warranted in systematically examining these differences and education practices within refugee and government schools that might influence these data.\nTeachers are central to the quality of refugee education in Kenya, as connected to issues of curriculum, pedagogy, and language. Only 27.4 percent of primary teachers in Kakuma are trained, through a combination of national teacher training programs and other short-courses (UNHCR, 2017c). Teachers are required to use the national Kenyan curriculum in all schools in Kenya, including in camps. Mastery over this curriculum determines success on school leaving exams, however content specific to the context of Kenya can be challenging for refugees. For example, teachers of refugees describe uncertainty in teaching about cultures, historical periods, scientific concepts, and geographical features that are Kenya-specific and unfamiliar to refugees (Bellino & Dryden-Peterson, Under review; Mendenhall et al., 2015). In some instances, teachers adapt the curriculum by encouraging students to share stories from their parents and countries of origin. In this sense, the teacher hopes to go beyond the national scope, connecting learning to a student\u2019s lived experiences in order to help students make meaning of the content (Bellino & Dryden-Peterson, 2014). Pedagogy in refugee contexts in Kenya, as in Kenya more broadly and in most developing contexts, is \u201ca neglected priority\u201d (Schweisfurth, 2015, p. 259). Mendenhall et al. (2015) found a dominance of teacher-centered practices, including recitation and lecture. Overcrowded classrooms can hinder teachers from implementing effective classroom management techniques. Yet there is variability of pedagogical practice across classrooms. Some teachers create charts and hands-on learning tools using locally available resources (Mendenhall et al., 2015). There is insufficient evidence to determine how schools are addressing the needs of students with learning and physical disabilities or students needing mental health counseling.\n28\nLanguage can be a barrier to refugee students accessing schools in Kenya. The official Kenyan language policy provides home language instruction in lower primary grades and English instruction in upper primary and secondary grades. However, in camp settings, refugee children start learning both Kiswahili and English upon enrollment. Kiswahili is a significant challenge among South Sudanese and Somali children, who arrive at the camps with divergent linguistic backgrounds, while those from Burundi and DRC have prior knowledge of Kiswahili, but lower proficiency in English. One teacher describes that \u201cThe students are\u2026 frustrated because they want to be with their age mates,\u201d but they do not have the English skills or the formal education to support their entry into a class where they are matched with peers (Mendenhall et al., 2015). In the Nairobi shared space schools, teachers had Kenyan certification and demonstrated competence in English and Kiswahili. However, teachers at the geographically separate camp-based schools were primarily refugees who spoke many languages and taught exclusively in English, in which they had varying degrees of competence. Notably, despite the linguistic heterogeneity in their classrooms, the teachers interviewed described their lack of training in how to incorporate origin home language instruction and support the acquisition of Kenyan languages of instruction (Mendenhall et al., 2015).\nBELONGING AND COHESION\nThe \u201cstructural integration\u201d of refugees in the Kenyan education system has not translated into \u201crelational integration\u201d of refugees. The experience of exclusion among refugees varies by age, role, nationality, and learning contexts, yet across the data that we analyzed, the common experience is a yearning for a sense of belonging. Refugees expressed challenges to forming meaningful relationships in the presence of stigma, discrimination, and physical harassment, and in the absence of formal recognition, protection and opportunities for upward mobility.\nIn camp settings, refugee children and youth report struggling to foster closer relationships with their teachers because of hardships such as malnourishment, long walks to school, limited resources at home, and psychosocial stress. Teachers\u2019 attempts to address students\u2019 psychological challenges comes with limited professional trainings and structural support. A certain level of mutual understanding between refugee and host community youth does however exist (Bellino & Dryden-Peterson, Under review). Kenyan students who enroll in camp schools in Kakuma are primarily Turkanas, a historically marginalized tribe within Kenya who experience similar struggles to refugee students. The shared experience of\n29\nmarginality thinly unites refugee and host community youth. That said, the extent to which this shared struggle translates into a sense of belonging for either group is a question that needs further research.\nRefugee students who are able to pursue educational opportunities outside of refugee camps face a different kind of exclusion, such as stigma and discrimination. Experiences of stigma often lead to identity-based discrimination inside and outside the classroom, resulting in name-calling, verbal abuse, and physical bullying. Somali students in particular seem to experience targeted discrimination because of their religious background and the political climate. Some Somali students cite that they have been called \u201cterrorists\u201d or \u201cAl Shabaab\u201d from teachers and Kenyan national students (Bellino & Dryden-Peterson, Under review). A common strategy for refugees to avoid identity-based discrimination is to hide their refugee status.\nThe experiences of stigmatization, routine discrimination, and marginalization amongst refugees persist, if not worsen, as refugee children leave the Kenyan education system and enter the workforce. The experience of belonging differs by decisions refugees make after secondary school. Those who choose to return to the camps for employment opportunities experience lack of formal recognition of their previous credentials. For example, refugee teachers are forced to subscribe to a power hierarchy within the camp schools despite having successfully navigated selective secondary schools.\nBox 3: Agency Education Staff in Kenya (using data derived from Bellino & Dryden-Peterson, 2014)\nLouisa, an education officer for an NGO working in Kakuma, is overworked but deeply committed to widening access to quality education for refugee children in Kenya. Louisa expresses, \u201cwe all care about education\u2026We have to make sure that we really create access. We don\u2019t block anyone from joining our schools...they are not denied.\u201d Louisa notes the main challenges to quality involve qualified staff, available teacher training, and overcrowded classrooms. Hiring a national Head Teacher with experience in government-run schools is preferable to hiring a refugee teacher with little to no experience. For teacher training, \u201c[Our NGO] offers 5-day induction courses... [and although] the challenges they [teachers] face are enormous\u201d, Louisa hopes \u201c... to guide them to do the right thing at the right time.\u201d Notably, congested classrooms are a concern. \u201cTeachers feel it is a big challenge; they cannot reach out to every learner\u2026\u201d\n30\nDue to these challenges in camp schools, Louisa believes it is critical for refugee students to access government schools. She states \u201c...we encourage them to go [to government-run schools]...we cannot accommodate everyone.\u201d In order to enable more students to advance their studies, Louisa sees promise in leveraging partnerships to assist students in accessing government-run schools.\nWhether in camp or government-run schools, Louisa emphasizes the importance of schools as safe spaces for children. She hopes to create and encourage connections to communities through involvement in peace education, clubs, and community dialogues. Connecting with others and normalcy for children are important elements of refugee education. At school, \u201cThey [the students] are good friends...at the end of the day, there is no discrimination\u2026school is a place that makes them be one...be it you\u2019re a Congolese, you\u2019re a Somali, you\u2019re a Sudanese, you are all one.\u201d\n5c. Lebanon\nCONTEXT\nConflict in Syria, which erupted in early 2011, has had debilitating impacts on Syria and neighboring countries hosting Syrian refugees. Conflict has claimed the lives of 470,000 Syrians and forced 10.9 million Syrians to flee their homes. 4.8 million have crossed into neighboring countries to seek refuge (Human Rights Watch, 2017). Lebanon is not a signatory to the 1951 Refugee Convention, nor to the 1967 Protocol. Displaced populations, including Syrians, are referred to either as \u201cdisplaced\u201d or \u201cnon-Lebanese\u201d in official discourse, and therefore Syrians must annually renew their residency permits to access social services such as education.\nFurther, in May 2015, Lebanon officially closed its borders to Syrian refugees and instructed UNHCR to stop registering Syrians, thereby curtailing their access to humanitarian services and rendering access to national services more difficult (Janmyr, 2016; Mendenhall, Russell, & Bruckner, 2017). The lack of ability to register also makes it challenging to determine the number of Syrian refugees in Lebanon. As of early 2018, UNHCR reported that Lebanon was host to 1,001,051 Syrian refugees, the second largest number of Syrian refugees in the region after Turkey, and the greatest number of refugees per capita worldwide (UNHCR, 2017f). Lebanon and Syria have a long history of complex and challenging political, economic and\n31\nsocial relations. Syria occupied Lebanon between 1976 and 2005; in 2005, Syrian troops officially withdrew from Lebanon, which has since continued to recover from war, including managing a precarious sectarian balance between its Sunni, Shia, Christian, and Druze groups (Blanchard, 2012). The Lebanese government and its citizens fear that an influx of Syrian refugees, predominantly Sunni, will disturb the country\u2019s delicate balance, thereby reigniting conflict dynamics within Lebanon. These tensions have important implications for the processes underlying the inclusion of Syrian refugees into Lebanese national systems, including public education.9\nIn response to the Syrian refugee crisis, in 2014 the Ministry of Education and Higher Education (MEHE) created the Reaching All Children with Education (RACE) strategy, a three-year strategy aimed to improve Syrians\u2019 access to quality education and to strengthen Lebanon\u2019s existing education systems to accommodate an influx of Syrian learners (Lebanon Ministry of Education and Higher Education, 2014). The following year MEHE set up a Program Management Unit (PMU) to manage educational policy and practices and ensure that all activities were streamlined under it, including coordination with multilateral and bilateral donors. In practice though, much focus centered not on quality but on increasing access to education for Syrians as well as for national students, across sectarian lines. In 2016, MEHE launched RACE II, with a revised agenda with an explicit focus on improving the overall quality of public education in Lebanon for refugee and national students alike (Lebanon Ministry of Education and Higher Education, 2016).\nAt present, Lebanon\u2019s education system is multilingual and characterized by a variety of school types. Languages of instruction include Arabic, English, and French, depending on the subject, grade-level, and geographic location (Vlaardingerbroek & Shehab, 2012). Three types of schooling are available for refugee students: formal public education, MEHE directed non-formal education programs (NFE), and private schools (Mendenhall et al., 2017). MEHE developed the NFE framework for Syrian refugee students in order to provide regulated programing for refugee learners who were unprepared to enter the public education system directly. Within this framework, there are a number of programs, the largest of which\n9 We acknowledge the long-standing presence of Palestinians in Lebanon and the need to examine their education. We do not include education of Palestinian refugees in Lebanon in the context of this analysis since we are focused on inclusion of refugees in national education systems and Palestinians attend separate schools managed by UNRWA in Lebanon.\n32\nis the Accelerated Learning Program (ALP). ALP\u2019s goal is to focus on students whose primary level educational attainment is below grade level and to accelerate their mainstreaming into the Lebanese public system. Non-formal education programs were initially acceptable in Lebanon and were operated by various other actors including national and international non-government organizations, religious groups, as well as outside governments and non-state actors involved in the Syrian conflict. However, since RACE was created, MEHE has increasingly called for the closure of any school operating outside its purview to ensure consistency in the types and quality of education Syrians can access (Buckner, Spencer, & Cha, 2017). Nevertheless, a number of these schools continue to operate, some even clandestinely, and serve refugee students unable or unwilling to enroll in government schools.\nACCESS\nRefugees can access several types of educational settings in Lebanon: government schools and government NFE programs, private and semi-subsidized schools, and refugee-only non-formal schools. Given our focus on the inclusion of refugees in national education systems, we focus on government schools. Since 2013, MEHE has permitted refugees legal and formal access to government schools at the primary level and to secondary school with proof of primary completion in Lebanon or Syria. With the onset of a defined strategy like RACE, to accommodate a growing population of Syrian learners, Lebanese students studied in the morning and Syrian students were allocated the same physical space to learn in the afternoon, often called the \u201csecond shift.\u201d The number of government schools providing a second shift has increased each year since the start of the program in 2013; as of the 2017/2018 school year, 348 government schools provided a second shift to refugee students. In these government schools, Syrian students follow the Lebanese curriculum, are taught by Lebanese teachers in languages of instruction followed in Lebanese government schools, and have access to the same certifications as Lebanese nationals (see Table 1). The model of inclusion used in Lebanon is primarily what we call temporally separate schools. Nationals and refugees attend the same schools but at different times (see Table 2).\nThere are numerous barriers and challenges for refugees in accessing education. Of the 500,000 Syrian children registered in Lebanon between the ages of 3 to 17 years, more than half of them are out-of-school with dropout rates as high 78 percent in some places such as the Bekaa valley (Government of Lebanon & United Nations, 2017). The main barriers include the costs associated with accessing education\n33\nand the limited capacity of the government school system, especially in rural areas with high numbers of refugees. While school fees are currently covered by international donors, parents are still required to pay for transportation to send their children to school, a cost which many families cannot often cover. The vast majority of Syrian families who are unable to procure their legal residency in Lebanon fear arrests and harassment at Lebanon\u2019s military checkpoints. For this reason Syrian refugee families, especially in rural areas, rely on their children instead to supplement or provide the household income and can therefore remain reluctant to enroll their children in school (Government of Lebanon & United Nations, 2017). It is in these instances that a family unit\u2019s legal status, as established by the Ministry of Interior, severely impacts refugee children\u2019s opportunities to access education, a public good governed by Lebanon\u2019s Ministry of Education.\nLack of access to education for refugees is further exacerbated by resource and capacity constraints of the government school system in Lebanon. Prior to the Syrian refugee crisis the government school system was already underfunded and accessed only by the most marginalized Lebanese nationals. Seventy percent of Lebanese students enroll in private or semi-subsidized schools, as the quality of government schools is considered to be inferior and Lebanese nationals with the ability to opt out of the public schools do so (CERD, 2016; Chami, 2016). In 2015-2016, the number of Syrian refugees enrolled in government primary and secondary schools (221,622) actually surpassed that of Lebanese children (249,494) (Human Rights Watch, 2016; Lebanon Ministry of Education and Higher Education, 2017a). This massive increase in demand for government schooling by refugees has led to a shortage of adequate school buildings, limited availability of textbooks, and gaps in appropriate levels of staffing, particularly for second shifts. Often, second shift teachers also teach during the first shift, which has implications for the sustainability of their work and the effectiveness of their teaching (Adelman & Dryden-Peterson, Under review).\nQUALITY\nThe focus of RACE I was on improving access to primary and secondary school for Syrians in Lebanon. RACE II placed greater emphasis on strengthening the national education system to eventually improve the quality of education available to both, refugees and nationals. To strengthen Lebanon\u2019s infrastructure and teacher capacity, RACE II included provisions to increase educational spaces to alleviate overcrowding, provide pedagogical training to teachers to enhance quality of instruction, undertake wide-scale curricular revisions, and implement an Education Management and Information System (EMIS).\n34\nWhile RACE II mandated the construction of additional schools, the second shift policy remains the primary strategy to address issues related to overcrowded classrooms. This model presents challenges for teachers and thus for quality. In a recent study (2014-2017) focused on the processes of structural and relational integration at three Lebanese government schools, teachers, especially those working first and second shifts, reported experiencing exhaustion and anxiety, and feeling overworked and under-regarded. Moreover, delays in teachers\u2019 payments have compounded these anxieties, fostering resentment and decreasing motivation amongst teachers, which has prompted greater teacher turnover in the second shift and teacher strikes (Adelman & Dryden-Peterson, Under review).\nTeachers working in the second shift not only face pedagogical challenges related to larger classes, but must also address the large diversity in students\u2019 complex educational needs. While there is minimal literature documenting pedagogical strategies of Lebanese teachers in government schools for either the first or second shift (Bahou, 2016; Lebanon Ministry of Education and Higher Education, 2016; Mattar, 2012), recent research suggests that teachers have yet to significantly adapt their pedagogy to meet the varied needs of their refugee students. In both the first and second shift, teachers tend to rely on teacher-centered strategies (e.g., rote-based learning, lecturing, etc.) with minimal student-teacher engagement, and as classroom sizes swell, teachers rely even more heavily on these strategies (Adelman & Dryden-Peterson, Under review). Research into classroom management is similarly sparse: while there are examples of teachers who utilize more positive disciplinary techniques, it appears that the use of physical and/or verbal threats is a persistent issue in Lebanese government schools, amongst both refugee and national students (Adelman & Dryden-Peterson, Under review). In addition, the multilingual nature of Lebanon\u2019s national curriculum makes inclusion in government schools more challenging for Syrian refugee children, who were previously taught in Arabic in Syria. Publicly available disaggregated data on refugee and Lebanese citizens\u2019 learning outcomes is not yet available.\nEven though teacher training and professional development could help to address several pedagogical challenges, it is not widely availabile to teachers (Adelman & Dryden-Peterson, Under review). As of July 2017, only 365 individuals out of a targeted 20,333 teachers and education personnel had participated in any type of official training regarding the inclusion of refugee students in government schools (United Nations, 2017). There is early evidence that even without any formal training, some teachers adapt the curricula in order to best serve their students. For example, teachers may explain key concepts in Arabic when a subject should be taught in English or French or slow their pace of teaching to allow students more\n35\ntime to learn in an unfamiliar language. Teachers may deviate from the official curriculum, summarizing general ideas over specific ones, or prioritize certain concepts within the curriculum while deliberately overlooking other areas, in an effort to move through the material at an adequate pace. However, these decisions are made ad hoc and are not systematic throughout the educational system (Adelman & Dryden-Peterson, Under review). Though these may create inconsistencies in the quality of education available to Syrian learners, these concessions may indirectly facilitate relationship-building between refugee learners and their teachers.\nBELONGING AND COHESION\nThough in principle, teachers are encouraged to treat Syrian children as they do Lebanese children, an approach which may foster relational inclusion, this expectation is rarely achieved in practice given the multiple, complex challenges of including Syrian refugee students in government schools (Adelman & Dryden-Peterson, Under review). Teachers reported observing violence between Lebanese and Syrian students in the schoolyard before and between shift changes; and some teachers recalled how Lebanese students would blame Syrian children when their belongings went missing or school desk or chairs were damaged. While teachers did their best to manage tensions between Syrian and Lebanese students, staff reported feeling unprepared to properly address these relational issues (Adelman & Dryden-Peterson, Under review). One study found that Syrian and Lebanese children in mixed first shift classes, the shared space model, had more positive perceptions of each other and stronger relationships with each other, than did Lebanese and Syrian students attending temporally separate schools (Abla & Al-Masri, 2015).\nSyrian children may also struggle to develop a sense of belonging in Lebanon. Once in Lebanon, Syrian refugee children may be physically safe from conflict, but recent research documents that they do not experience psychological safety in their communities, or even in their schools. A situation analysis of Syrian youth in Lebanon found that 81 percent of Syrian females between 15 and 18 years reported having no Lebanese friends and 59 percent of those sampled in this age group did not feel safe in the country (UNFPA, UNESCO, UNICEF, UNHCR, & Save the Children, 2014, p. 33). Additionally, Syrian children\u2019s pre-displacement experiences including their exposure to violence and loss of familial relationships, have profound bearings on their psychosocial wellbeing when displaced to neighboring countries, including Lebanon (Save the Children, 2017).\n36\nAt present, limited provisions are made to provide psychosocial support to Syrian refugee students in Lebanon, constraining the long-term potential to cultivate a sense of belonging through educational spaces. MEHE has tried to address the needs of children broadly by employing a psychosocial support counselor at each school to provide general support to refugees\u2019 well-being at school, and by piloting the inclusion of a psychosocial support component in its accelerated learning program (International Alert, 2015). While more targeted efforts towards building positive relationships between Syrians and Lebanese are needed, psychosocial interventions may serve as one of several mechanisms for aiding refugee students\u2019 transition into the Lebanese education system.\nThough the relational experiences of Syrian learners in Lebanon\u2019s schools are mixed, \u201cstructural integration\u201d of Syrians into government schools can be an important, first step in working toward \u201crelational integration.\u201d This \u201cstructural integration\u201d at one level of education connects to possible \u201cstructural integration\u201d and \u201crelational integration\u201d at other educational levels and vis-\u00e0-vis livelihoods. For instance, in 2017-2018, there were only 3,938 non-Lebanese students enrolled in secondary schools (Lebanon Ministry of Education and Higher Education, 2017a). For the select few who complete a full cycle of school education, access to and experiences of tertiary education are uneven. Furthermore, opportunities for formal legal employment for Syrians, even when educated in Lebanon, are limited to just three sectors including agriculture, construction, and the environment (Chopra, 2018). These factors shed light on the ways unequal structures may circumscribe refugee children\u2019s academic and personal goals, eroding senses of belonging and cohesion.\nBox 2. National Teacher of Refugees in Lebanon (using data derived from Adelman & Dryden-Peterson, Under review)\nHiba is proud to serve as principal at one of the many government schools selected to initiate Lebanon\u2019s second shift policy. It is an exhausting job\u2014she works 12 hour days through both shifts and though she oversees double the number of students and staff than originally trained and equipped to do. However Hiba sees no other means to ensuring that all of her students, regardless of their national identity, obtain an education. In addition to focusing on the academic progress of her students, Hiba is\n37\ncommitted to creating a positive school culture where children feel a sense of belonging. She encourages her teachers to show the students care and love. She explains, refugees students \u201cneed the proper care\u201d as they are currently suffering difficult circumstances. Hiba encourages her teachers to \u201clove them as well as your kids\u2026 If you don\u2019t love the Syrian students you don\u2019t love anything.\u201d\nDespite her efforts, Hiba knows her school has a long way to go in fostering social cohesion and addressing the socio-emotional needs of children who have experienced trauma and displacement. She has witnessed the effects the conflict in Syria has had on children in the ways they interact with each other. She describes how at recess, refugee children tend to interact violently with each other, as if \u201cthey are fighting an enemy.\u201d Hiba tries to encourage greater cohesion among Syrian students, reminding them that they \u201care from one home country [and they should] like each other, love each other.\u201d She sees promoting cohesion between Syrian and Lebanese students enrolled in her school as just as essential, regardless of the social and political tensions brewing outside her school\u2019s walls. In particular, she draws a distinction between the citizens of Syria and the state of Syria, recognizing that \u201cLebanon doesn\u2019t like the government of Syria, but what about the kids? What is the relationship between the government and the kids?\u201d Hiba shakes her head, \u201clet them learn.\u201d\n38\n6. CHALLENGES AND PROMISING PRACTICES FOR INCLUSION OF REFUGEES IN NATIONAL EDUCATION SYSTEMS\nAmong the three country case studies, efforts to enhance access, quality, and sense of belonging and cohesion in host country education systems for refugee students point to several promising practices. For readability and clarity, Table 3 is divided into three categories\u2014access, quality, and belonging; however, as stated earlier in this Background Paper, these categories are mutually dependent and interrelated. Supporting teachers was identified as essential for implementing the most promising practices and improving refugee students\u2019 learning outcomes. Teachers benefit from cultural, curricular, and pedagogical bridging strategies that relate to refugee students\u2019 backgrounds within educational systems. We also recognize that some of these teacher practices may not be systematized, mandated, or even scripted within national policy. Nonetheless teacher practices that are deviations from the norm provide an important perspective to examine the experiences of refugee learners within government school classrooms. Refugee students fleeing violence often struggle with psychosocial stress created as a consequence of conflict; providing these students with psychosocial support is vital to improving their learning outcomes and fostering cohesion and a sense of belonging within their new communities. Efficient stakeholder collaboration enabled relevant actors to provide comprehensive services.\nTable 3. A Selection of Promising Practices that Act on Key Challenges in Inclusion of Refugees in National Education Systems\nPriority Challenge Promising Practice(s)\nCoordination between stakeholders\nEstablished trust between global, national and local actors over long history of partnership; regular site-based meetings between UNHCR and local stakeholders (Bellino & Dryden-Peterson, 2014)\n39\nAccess\nOpportunity cost of attending school\nFull or partial subsidies to refugee and host community families to attend school (Lebanon Ministry of Education and Higher Education, 2017b)\nAccess to secondary education\nScholarships to attend government schools (Bellino & Dryden-Peterson, 2014; Buckner & Spencer, 2016); second shift/ two schools in one (Murwanjama & Mureu, 2017)\nLegal frameworks that protect the right to education\nRight to movement; right to attend government schools (Government of Uganda et al., 2017; Lebanon Ministry of Education and Higher Education, 2014, 2016)\nCoordination between stakeholders and community\nCommunity mobilizers who do home visits when there are extended absences or other family issues (Bellino & Dryden-Peterson, Under review)\nTeacher training and development\nOngoing pre-service and in-service professional development; prepare teachers for the particular challenges that refugee students and communities may face (Bellino & Dryden-Peterson, 2014; Mendenhall, 2017)\nLanguage(s) of instruction\nAccelerated learning programs (ALP); multilingual classrooms (Accelerated Education Working Group, 2017; Reddick & Dryden-Peterson, 2018)\nStudent-to-teacher ratio\nSmaller class sizes allow for the implementation of student-centered pedagogies (Bellino & Dryden-Peterson, 2014)\n40\nQuality\nClassroom resources\nTeacher-created free- or low-cost local materials: posters, organic materials, reused items (Mendenhall et al., 2015)\nBelonging\nStudents social-emotional well-being and capacity to learn\nImplementation of psycho-social support (e.g., counseling, integration of psycho-social/socio-emotional class) (Norwegian Refugee Council, 2017; Tubbs & Weiss-Yagoda, 2017)\nSocial cohesion between national and refugee populations\nFoster environment of empathy through storytelling in shared spaces; parent-teacher engagement (e.g., home visits, parent involvement in the classroom) (Dryden-Peterson, 2017; Reddick & Dryden-Peterson, 2018)\nCreating a learning experience relevant to refugees\u2019 experiences\nCultural bridging and facilitating belonging whereby teachers understand the relationship between curriculum and students; building relationships (Bellino & Dryden-Peterson, 2014)\n41\n7. SYNTHESIS: INCLUSION OF REFUGEES IN NATIONAL EDUCATION SYSTEMS\nThe Comprehensive Refugee Response Framework (CRRF) advocates for policies that can facilitate the inclusion of refugees into national systems in order to create conditions that are productive for both refugees and nationals (United Nations General Assembly, 2016). Moreover, SDG 16: Peace, Justice and Strong Institutions, calls for the promotion of inclusive societies and institutions (United Nations, 2016), a goal that implicitly supports the \u201crelational integration\u201d that represents an individual sense of belonging and social cohesion within shared communities. However, the case studies of Uganda, Kenya, and Lebanon demonstrate critical gaps between the \u201cstructural integration\u201d enabled by policies of inclusion and outcomes of \u201crelational integration\u201d within the educational experiences that refugee students have, modeled in Figure 1 below. In the Figure, we show that \u201cstructural integration,\u201d such as the use of the national curriculum, instruction by national teachers in the languages of the host country, and access to national examination and certification are prerequisites to possibilities of educational experiences that allow for processes of developing a sense of belonging and social cohesion, including productive everyday experiences of teaching and learning and relationship-building between refugees and nationals. Yet we observe in our empirical cases that \u201crelational integration\u201d does not automatically follow from \u201cstructural integration\u201d policies but requires concerted, systematic effort focused on educational experiences and processes of developing a sense of belonging and social cohesion.\nWithin each model of inclusion, we examined the interrelated dimensions of access, quality, and belonging and cohesion. It is through these dimensions that the connections between structural and relational integration are illuminated. In the cases of Uganda, Kenya, and Lebanon, each country has adopted official policies of inclusion that enable \u201cstructural integration\u201d; however, different geographic and social realities of refugee and national communities have limited the ways in which these policies create conditions for \u201crelational integration.\u201d For example, refugee and national families in Uganda and Kenya are often geographically separate in where they live, which results in a geographically separate model of education, even under policies of inclusion. Further, Lebanese students initially shared classrooms with refugees, but when class size and social tensions became untenable, a second shift policy was adopted, resulting in a temporally separate model of education. These different models of practice\n42\nfor policies of inclusion circumscribe the kinds of education students have access to, the quality of that education, and the nature of relationship-building that can happen between refugees and nationals.\nWe also observe that the substance of everyday teaching and learning and relationships have implications for possibilities for productive processes to develop a sense of belonging and social cohesion. For example, while refugee and national students in Kenya and Uganda at times shared space in the same schools, allowing for some degree of relationship-building, national curricular and language policies limited individual refugee students\u2019 capacities to feel represented in the classroom. Similarly, in Lebanon temporally separate situations limited interactions between Syrian refugee and Lebanese students to transitional periods during the school day. While this policy of separate shifts was in part intended to mitigate interpersonal conflict between students, it also has limited opportunities for refugee and national students to develop relationships with each other, and to cultivate an individual sense of belonging and collective social cohesion.\nPolicies and associated practices designed to include refugees in national systems of education need to be analyzed and measured by the extent to which they enable both \u201cstructural integration\u201d and \u201crelational integration.\u201d A country that adopts a model of inclusion that focuses on both has the unique opportunity to improve the welfare of both nationals and refugee and to build toward strengthening national institutions that reflect the goals of social cohesion and inclusive communities embodied in the global goals of both the Sustainable Development Goals and the Comprehensive Refugee Response Framework.\n43\nFigure 1: Inclusion: Connecting policies of \u201cstructural intregration\u201d to outcomes of \u201crelational integration\u201d\nPolicies to include refugee students in national education systems can create the conditions for \u201cstructural integration.\u201d The nature of educational experiences that students have on a daily basis \u2013 comprising both the teaching and learning practices and the relationships \u2013 are central to the processes of developing a sense of belonging and/or social cohesion. Through educational experiences that productively develop a sense of belonging and social cohesion, outcomes of \u201crelational integration\u201d can occur.\nPROCESSES of developing Sense of Belonging individual level for each student Social Cohesion among nationals and refugees POLICIES of \u201cStructural Integration\u201d Access shared or separate space models Quality use of national curriculum, national teachers, national certification OUTCOMES of \u201cRelational Integration\u201d Presence of deep, productive relationships between nationals and refugees Shared sense of membership among nationals and refugees within a learning community Educational EXPERIENCES Everyday teaching and learning practices within classrooms Everyday relationships among refugee and national students All students who access school have educational EXPERIENCES. The nature of the teaching and learning practices and the relationships within these experiences can be productive or destructive to the PROCESSES of developing a sense of belonging and/or social cohesion.\nREFERENCES\nAbla, Z., & Al-Masri, M. (2015). Better together: The impact of the schooling system of Lebanese and Syrian displaced pupils on social stability. Retrieved from London: https://data2.unhcr.org/en/documents/download/44275\nAbu El-Haj, T. R. (2015). Unsettled belonging : educating Palestinian American youth after 9/11. Chicago ; London: University of Chicago Press.\nAccelerated Education Working Group. (2017). The case for Accelerated Education. Retrieved from http://www.unhcr.org/59ce4f5b4\nAdelman, E., & Dryden-Peterson, S. (Under review). Reaching for all in Lebanon: A case study of the integration of Syrian refugee students into Lebanese public schools.\nAlbaugh, E. (2014). State-building and multilingual education in Africa. NYC, New York: Cambridge University Press.\nAllport, G. (1979 [1954]). The nature of prejudice. Reading, MA: Addison-Wesley Publishing Company.\nAltinyelken, H. K., Moorcroft, S., & van Der Draai, H. (2014). The dilemmas and complexities of implementing language-in-education policies: Perspectives from urban and rural contexts in Uganda. International Journal of Educational Development, 36, 90-99. doi:10.1016/j.ijedudev.2013.11.001\nAntonsich, M. (2010). Searching for Belonging - An Analytical Framework. Geography Compass, 4(6), 644-659. doi:doi:10.1111/j.1749-8198.2009.00317.x\nBahou, L. (2016). \u2018Why do they make us feel like we\u2019re nothing? They are supposed to be teaching us to be something, to even surpass them!\u2019: Student (dis)engagement and public schooling in conflict-affected Lebanon. Cambridge Journal of Education, 1-20. doi:10.1080/0305764X.2016.1216086\nBajaj, M., & Bartlett, L. (2017). Critical transnational curriculum for immigrant and refugee students. Curriculum Inquiry, 47(1), 25-35. doi:10.1080/03626784.2016.1254499\nBanks, J. A. (Forthcoming). Global Migration, Failed Citizenship, and Civic Education. In M. S. Soronso & M. Su\u00e1rez-Orozco (Eds.), On Humanitarianism and Mass Migration. Los Angeles; New York: University of California Press; Ross Institute of New York.\n45\nBar-Tal, D., & Rosen, Y. (2009). Peace Education in Societies Involved in Intractable Conflicts: Direct and Indirect Models. Review of Educational Research, 79(2), 557-575.\nBarbelet, V., Hagen-Zanker, J., & Mansour-Ille, D. (2018). The Jordan Compact: Lessons learnt and implications for future refugee compacts. Retrieved from London:\nBartlett, L., & Garcia, O. (2011). Additive schooling in subtractive times : bilingual education and Dominican immigrant youth in the Heights. Nashville.\nBellino, M. J. (2017). Youth in postwar Guatemala : education and civic identity in transition. New Brunswick, NJ: Rutgers University Press.\nBellino, M. J., & Dryden-Peterson, S. (2014). Implementing a Global Strategy in National Contexts: Field-based case studies of the UNHCR Education Strategy, 2012-2016, Kenya Case Study. Retrieved from Unpublished:\nBellino, M. J., & Dryden-Peterson, S. (Under review). Inclusion and Exclusion within a Policy of National Integration: Refugee Education in Kenya\u2019s Kakuma Refugee Camp.\nBellino, M. J., & Williams, J. H. (2017). (Re)Constructing Memory: Education, Identity, and Conflict. Rotterdam.\nBenson, C. (2012). The Role of Language Instruction in Promoting Quality and Equity in Primary Education. . In J. H. A. Cassola (Ed.), Lessons in Educational Equality (pp. 191-221). Oxford, UK: Oxford University Press.\nBetts, A., & Collier, P. (2017). Refuge : rethinking refugee policy in a changing world. New York, NY: Oxford University Press.\nBlanchard, C. (2012). Lebanon: Background and U.S. Policy. Retrieved from\nBuckner, E., & Spencer, D. (2016). Educating Syrian Refugees in Lebanon. Retrieved from http://carnegieendowment.org/sada/63513\nBuckner, E., Spencer, D., & Cha, J. (2017). Between Policy and Practice: The Education of Syrian Refugees in Lebanon. Journal of Refugee Studies, fex027-fex027. doi:10.1093/jrs/fex027\n46\nBurns, A. (2010). Feeling the pinch: Kenya, Al-Shabaab, and East Africa's refugee crisis. Refuge, 27(1), 5-15.\nCampbell, E. H. (2006). Urban refugees in Nairobi: Problems of protection, mechanisms of survival, and possibilities for integration. Journal of Refugee Studies, 19(3), 396-413.\nCederman, L.-E., Weidmann, N. B., & Gleditsch, K. S. (2011). Horizontal Inequalities and Ethnonationalist Civil War: A Global Comparison. The American Political Science Review, 105(3), 478-495. doi:http://dx.doi.org/10.1017/S0003055411000207\nCERD. (2016). Statistical Bulletin for the Academic Year. In. Beirut: CERD.\nChami, M. (2016, October 12). Public schooling in Lebanon: Difficult conditions but can still be engaged. The Daily Star. Retrieved from http://www.dailystar.com.lb/Business/Local/2016/Oct-12/376114-public-schooling-in-lebanon-difficult-conditions-but-can-still-be-changed.ashx\nChopra, V. (2018). Learning to Belong, Belonging to Learn: Syrian Refugee Youths' Pursuits of Education, Membership and Stability in Lebanon. (EdD), Harvard University, Cambridge, MA.\nChopra, V., & Adelman, E. (2017). The Pursuit, Practicality and Potential of Refugee Education. ZEP : Zeitschrift f\u00fcr Internationale Bildungsforschung und Entwicklungsp\u00e4dagogik, 40(1), 4-9.\nCrawford, N., Cosgrave, J., Haysom, S., & Walicki, N. (2015). Protracted displacement: Uncertain paths to self-reliance in exile. Retrieved from London: http://www.odi.org/sites/odi.org.uk/files/odi-assets/publications-opinion-files/9851.pdf\nCummins, J. (1981). Empirical and Theoretical Underpinnings of Bilingual Education. Journal of Education, 163(1), 16-29. doi:10.1177/002205748116300104\nCummins, J. (2000). Language, power, and pedagogy : bilingual children in the crossfire. Buffalo, N.Y.: Clevedon.\nDevictor, X., & Do, Q.-T. (2016). How Many Years Have Refugees Been in Exile? Retrieved from Washington, DC:\n47\nDryden-Peterson, S. (2006). The Present is Local, the Future is Global? Reconciling Current and Future Livelihood Strategies in the Education of Congolese Refugees in Uganda. Refugee Survey Quarterly, 25, 81-92.\nDryden-Peterson, S. (2011a). Refugee Children Aspiring toward the Future: Linking Education and Livelihoods. In K. Mundy & S. Dryden-Peterson (Eds.), Educating Children in Conflict Zones: Research, Policy, and Practice for Systemic Change (A Tribute to Jackie Kirk) (pp. 85-99). New York: Teachers College Press.\nDryden-Peterson, S. (2011b). Refugee Education: A Global Review. Retrieved from Geneva:\nDryden-Peterson, S. (2015). Refugee education in countries of first asylum: Breaking open the black box of pre-resettlement experiences. Theory and Research in Education, 1-18.\nDryden-Peterson, S. (2016). Refugee Education: The Crossroads of Globalization. Educational Researcher, 45(9), 473-482.\nDryden-Peterson, S. (2017). Refugee education: Education for an unknowable future. Curriculum Inquiry, 47(1), 14-24. doi:10.1080/03626784.2016.1255935\nDryden-Peterson, S., Adelman, E., Bellino, M., & Chopra, V. (Under review). The Purposes of Refugee Education: Policy and Practice of Integrating Refugees into National Education Systems.\nDryden-Peterson, S., Dahya, N., & Adelman, E. (2017). Pathways to educational success among refugees: Connecting local and global resources. American Educational Research Journal, 54(6), 1011-1047.\nDryden-Peterson, S., & Hovil, L. (2004). A Remaining Hope for Durable Solutions: Local Integration of Refugees and their Hosts in the Case of Uganda. Refuge, 22, 26-38.\nDryden-Peterson, S., & Sieb\u00f6rger, R. (2006). Teachers as memory makers: Testimony in the making of a new history in South Africa. International Journal of Educational Development, 26(4), 394-403.\nGanimian, A. J., & Murnane, R. J. (2016). Improving Education in Developing Countries: Lessons From Rigorous Impact Evaluations. Review of Educational Research, 86(3), 719-755. doi:10.3102/0034654315627499\n48\nGarcia, O. (2012). Ethnic identity and language policy. In B. Spolsky (Ed.), The Cambridge Handbook of Language Policy (pp. 79-99). Cambridge, England: Cambridge University Press.\nGeorge, A. L., & Bennett, A. (2005). Case studies and theory development in the social sciences. Cambridge, Mass.: MIT Press.\nGerring, J. (2004). What Is a Case Study and What Is It Good for? American Political Science Review, 98(2), 341-354.\nGlobal Education Monitoring Report. (2017a). Aid to education is stagnating and not going to countries most in need. Retrieved from Paris: http://unesdoc.unesco.org/images/0024/002495/249568e.pdf\nGlobal Education Monitoring Report. (2017b). Concept Note for the 2019 Global Education Monitoring Report on Education and Migration. Retrieved from Paris: https://en.unesco.org/gem-report/sites/gem-report/files/Concept%20Note%205%20April%20Final.pdf\nGovernment of Kenya. (2014). Press statement by Cabinet Secretary for Interior & Coordination of National Government on Refugees and National Security Issued on 26th March 2014. Retrieved from http://www.hrw.org/sites/default/files/related_material/PRESS%20STATEMENT%20BY%20CABINET%20SECRETARY%20FOR%20INTERIOR%20%20COORDINATION%20OF%20NATI%20%20%20.pdf\nGovernment of Lebanon, & United Nations. (2017). Lebanon crisis response plan: 2017-2020. Beirut, Lebanon: Government of Lebanon & United Nations.\nGovernment of Uganda, United Nations, & UNHCR. (2017). Requirements for a comprehensive refugee response in Uganda. Retrieved from Kampala, Uganda:\nHathaway, J. C. (2016). A global solution to a global refugee crisis. European papers: a journal on law and integration(1), 93-99.\n49\nHuman Rights Watch. (2016). Growing up without an education: Barriers to education for Syrian refugee children in Lebanon. Retrieved from https://www.hrw.org/sites/default/files/report_pdf/lebanon0716web.pdf\nHuman Rights Watch. (2017). Syria: Events of 2016. Retrieved from New York: https://www.hrw.org/world-report/2017/country-chapters/syria\nIGAD Member States. (2017). Djibouti Declaration on Regional Conference on Refugee Education in IGAD Member States, Theme: Regional Quality Education Srandards and Inclusion into National Systems for refugee children in line with CRRF, SDG 4 and Agenda 2063 on Education. Retrieved from Djibouti:\nInternational Alert. (2015). Better Together: The Impact of the Schooling System of Lebanese adn Syrian Dispalced Pupils on Social Stability. Retrieved from London: http://civilsociety-centre.org/resource/better-together-impact-schooling-system-lebanese-and-syrian-displaced-pupils-social\nJanmyr, M. (2016). Precarity in Exile: The Legal Status of Syrian Refugees in Lebanon. Refugee Survey Quarterly, 35(4), 58-78.\nKorac, M. (2003). Integration and How We Facilitate It: A Comparative Study of the Settlement Experiences of Refugees in Italy and the Netherlands. Sociology, 37(51-68).\nLebanon Ministry of Education and Higher Education. (2014). Reaching all children with education in Lebanon. Lebanon: Ministry of Education and Higher Education.\nLebanon Ministry of Education and Higher Education. (2016). Reaching all children with education: RACE II (2017-2021). Lebanon: Ministry of Education and Higher Education.\nLebanon Ministry of Education and Higher Education. (2017a). RACE 2 Fact Sheet - November 2017. Retrieved from Beirut:\nLebanon Ministry of Education and Higher Education. (2017b). RACE Lebanon: Presentation to Education Partners Meeting. MEHE. Beirut, Lebanon.\n50\nLewin, K. M. (2009). Access to education in sub-Saharan Africa: patterns, problems and possibilities. Comparative Education, 45(2), 151-174.\nLloyd, C., Mete, C., & Sathar, Z. A. (2005). The Effect of Gender DIfferences in Primary School Access, Type, and Quality on the Decision to Enroll in Rural Pakistan. Economic Development and Cultural Change, 53(3), 685-710.\nMattar, D. M. (2012). Factors affecting the performance of public schools in Lebanon. International Journal of Educational Development, 32(2), 252-263. doi:http://dx.doi.org/10.1016/j.ijedudev.2011.04.001\nMendenhall, M. (2017). STRENGTHENING TEACHER PROFESSIONAL DEVELOPMENT: Local and global communities of practice in Kakuma Refugee Camp I Kenya. Retrieved from London:\nMendenhall, M., Dryden-Peterson, S., Bartlett, L., Ndirangu, C., Imonje, R., Gakunga, D., . . . Tangelder, M. (2015). Quality Education for Refugees in Kenya: Pedagogy in Urban Nairobi and Kakuma Refugee Camp Settings. Journal on Education in Emergencies, 1(1), 92-130.\nMendenhall, M., Russell, S. G., & Bruckner, E. (2017). Urban Refugee Education: Strengthening Policies and Practices for Access, Quality, and Inclusion.\nMilner, J., & Loescher, G. (2011). Responding to protracted refugee situations: Lessons from a decade of discussion. Retrieved from Oxford:\nMulumba, D. (2018, 11 January). Refugees in Uganda: An overview of the history, legal framework, challenges and opportunities. Paper presented at the \"Open Doors\" Refugee Symposium, Makerere University, Kampala, Uganda.\nMurwanjama, J., & Mureu, P. (2017). TWO SCHOOLS IN ONE: Management of high enrollment in refugee secondary schools. Retrieved from London:\nNational Curriculum Development Centre. (2018). Lower Primary/ Thematic Curriculum. Retrieved from Kampala: http://www.ncdc.go.ug/curriculum/lower-primary-thematic-curriculum\nNicholson, S. (2006). Accelerated Learning in Postconflict Settings: A Discussion Paper. Retrieved from info.worldbank.org/etools/library/latestversion.asp?241472\n51\nNicolai, S., Wild, L., Wales, J., Hine, S. b., & Engel, J. (2014). Unbalanced progress: What political dynamics mean for education access and quality. Retrieved from London:\nNorwegian Refugee Council. (2017). Non-formal education programming: enabling refugee children to enrol and integrate into the formal certified education system. Retrieved from https://www.promisingpractices.online/norwegian-refugee-council\nNovelli, M. (2016). Capital, inequality and education in conflict-affected contexts. British Journal of Sociology of Education, 37(6), 848-860. doi:10.1080/01425692.2016.1165087\nOECD. (2011). Social Cohesion in a Shifting World. Retrieved from Paris: http://www.oecd.org/site/devpgd2012/49067839.pdf\nPaulson, J. (2015). \"Whether and How?\" History Education about Recent and Ongoing Conflict: A Review of Reseach. Journal on Education in Emergencies, 1(1), 14-47.\nPavanello, S., Elhawary, S., & Pantuliano, S. (2010). Hidden and exposed: Urban refugees in Nairobi, Kenya. Retrieved from London:\nPiper, B., & Miksic, E. (2011). Mother tongue and reading: Using early grade reading assessments to investigate language-of-instruction policy in East Africa. In A. G. A. Wetterberg (Ed.), The early grade reading assessment: Applications and interventions to improve basic literacy (pp. 139-182). Research Triangle Park, NC: RTI Press.\nPlan International. (2017). Putting the CRRF into Practice: General Issues and Specific Considerations in Tanzania and Uganda. Retrieved from Tanzania; Uganda:\nReddick, C., & Dryden-Peterson, S. (2018). Language of Instruction in Refugee Education: A Review of Literature, Policies and Practices, A report prepared for the United Nations High Commissioner for Refugees.\nThe Refugees Act 2006, No. 47 Volume XCVIX, Act 21 Stat. (2006).\nReijerse, A., Van Acker, K., Vanbeselaere, N., Phalet, K., & Duriez, B. (2013). Beyond the Ethnic-Civic Dichotomy: Cultural Citizenship as a New Way of Excluding Immigrants. Political Psychology, 34(4), 611-630. doi:10.1111/j.1467-9221.2012.00920.x\n52\nRowley, E. A., Burnham, G. M., & Drabe, R. M. (2006). Protracted Refugee Situations: Parallel Health Systems and Planning for the Integration of Services. Journal of Refugee Studies, 19(2), 158-186. doi:10.1093/jrs/fej019\nRuiz, R. (1984). Orientations in language planning. NABE: The Journal for the National Association for Bilingual Education, 8(2), 15-34.\nSave the Children. (2016). Every Last Child: The children the world chooses to forget. Retrieved from London:\nSave the Children. (2017). Invisible Wounds: The impact of six years of war on the mental health of Syria\u2019s children. Retrieved from Australia: https://www.savethechildren.org.au/__data/assets/pdf_file/0007/196531/Invisible-Wounds.pdf\nSchweisfurth, M. (2015). Learner-centred pedagogy: Towards a post-2015 agenda for teaching and learning. International Journal of Educational Development, 40(1), 259-266.\nShin, S. J. (2013). Bilingualism in schools and society : language, identity, and policy. New York.\nSmagorinsky, P. (2008). The Method Section as Conceptual Epicenter in Constructing Social Science Research Reports. Written Communication, 25(3), 389-411.\nStewart, F., Brown, G. K., & Langer, A. (2008). Major Findings and Conclusions on the Relationship Between Horizontal Inequalities and Conflict. In F. Stewart (Ed.), Horizontal inequalities and conflict : understanding group violence in multiethnic societies (pp. 285-300). Basingstoke [England] ; New York: Palgrave Macmillan.\nStrang, A., & Ager, A. (2010). Refugee Integration: Emerging Trends and Remaining Agendas. Journal of Refugee Studies, 23(4), 589-607.\nTembe, J., & Norton, B. (2008). Promoting Local Languages in Ugandan Primary Schools: The Community as Stakeholder. Canadian Modern Language Review-Revue Canadienne Des Langues Vivantes, 65(1), 33-60. doi:10.3138/cmlr.65.1.33\n53\nTrudell, B. (2016). The impact of language policy and practice on children's learning: Evidence from Eastern and Southern Africa. Retrieved from Nairobi, Kenya:\nTubbs, C., & Weiss-Yagoda, J. (2017). AN EIE RESEARCH-PRACTICE PARTNERSHIP: Learning to improve academic and social-emotional outcomes. Retrieved from London:\nUNESCO. (2014). EFA Global Monitoring Report 2013-14, Teaching and Learning: Achieving Quality for All. Retrieved from Paris:\nUNESCO. (2016). Incheon Declaration and Framework for Action for the implementation of Sustainable Development Goal 4. Retrieved from Paris:\nUNESCO. (2017). National Teacher Policy: Qualified Teachers for Quality Education in Uganda. Retrieved from Kampala:\nUNESCO Institute of Statistics. (2018). Uganda. Retrieved from http://uis.unesco.org/country/UG\nUNFPA, UNESCO, UNICEF, UNHCR, & Save the Children. (2014). Situation Analysis of youth in Lebanon affected by the Syrian crisis. Retrieved from London: http://www.unfpa.org.lb/Documents/Situation-Analysis-of-the-Youth-in-Lebanon-Affecte.aspx\nUNHCR. (2011). The 1951 Refugee Convention: Reservations and Declarations. Retrieved from http://www.unhcr.org/cgi-bin/texis/vtx/search?page=&comid=4dac387c6&cid=49aea9390&keywords=reservations\nUNHCR. (2012). Education Strategy 2012-2016. Retrieved from Geneva:\nUNHCR. (2013). The UNHCR Uganda Strategy for Refugee Education, 2013-2016. Retrieved from Kampala:\nUNHCR. (2015). 2015 Global Child Protection, Education & SGBV Strategy Implementation Report. Retrieved from Geneva:\nUNHCR. (2016). Global Trends: Forced Displacement in 2015. Retrieved from Geneva:\nUNHCR. (2017a). 2016 Kenya Certificate of Primary Education (KCPE) Result Analysis and Transition Crunch. Retrieved from Kenya:\n54\nUNHCR. (2017b). 2017 Country operations profile - Kenya. Retrieved from http://www.unhcr.org/ke/\nUNHCR. (2017c). 2017 Teacher Numbers and Training Information - Kakuma Camps and Kalobeyei Settlement. Retrieved from\nUNHCR. (2017d). East, horn of Africa and the great leaks region refugees and asylum seekers by country of asylum as of 30 September 2017. Retrieved from https://data2.unhcr.org/es/documents/download/60678\nUNHCR. (2017e). Education dashboard - Kenya. Retrieved from http://www.unhcr.org/ke/\nUNHCR. (2017f). Global Trends: Forced Displacement in 2016. Retrieved from Geneva:\nUNHCR. (2017g). Left Behind: Refugee Education in Crisis. Retrieved from Geneva:\nUNICEF. (2016). The impact of language policy and practice on children's learning: Evidence from Eastern and Southern Africa. Retrieved from https://www.unicef.org/esaro/UNICEF(2016)LanguageandLearning-FullReport(SingleView).pdf\nUNICEF. (2017). South Sudan. Retrieved from https://www.unicef.org/southsudan/education.html\nUnited Nations. (2016). Sustainable Development Goals: 17 Goals to Transform Our World. Retrieved from http://www.un.org/sustainabledevelopment/education/\nUnited Nations. (2017). 2017 November Statistical Dashboard. Retrieved from Beirut, Lebanon:\nUnited Nations General Assembly. (2016). New York Declaration for Refugees and Migrants. Retrieved from New York:\nUnited Nations General Assembly. (2017). Report on the commission of the inquiry on Burundi. Retrieved from http://www.un.org/en/ga/search/view_doc.asp?symbol=A/HRC/36/54\nUnited Nations Security Council. (2017a). Report of the secretary-general on South Sudan (covering the period from 2 June to 1 September 2017). Retrieved from http://www.un.org/en/ga/search/view_doc.asp?symbol=S/2017/784\n55\nUnited Nations Security Council. (2017b). Report of the secretary-general on the United Nations organization stabilization mission in the Democratic Republic of the Congo. Retrieved from http://www.un.org/en/ga/search/view_doc.asp?symbol=S/2017/824\nVlaardingerbroek, B., & Shehab, S. S. (2012). Educational assessment in Lebanon. Assessment in Education: Principles, Policy & Practice, 19(3), 379-386. doi:10.1080/0969594X.2011.637900\nWard, M., Penny, A., & Read, T. (2006). Education Reform in Uganda - 1997 to 2004. Reflections on Policy, Partnership, Strategy and Implementation. Retrieved from https://assets.publishing.service.gov.uk/media/57a08c16ed915d3cfd001168/EducationPaperNo60.pdf\nWilson, N. J. (2014, December 30). Kenya's harsh new security laws put hundreds of thousands of refugees at risk. The Conversation. Retrieved from http://theconversation.com/kenyas-harsh- new-security- laws-put- hundreds-of- thousands-of-refugees-at- risk-35789\nWorld Bank. (2017). DataBank education statistics - Uganda. Retrieved from http://databank.worldbank.org/data/reports.aspx?source=Education%20Statistics#\nWorld Bank Group. (2016). Uganda's progressive approach to refugee management. Retrieved from Washington, DC: http://www.worldbank.org/en/topic/fragilityconflictviolence/brief/ugandas-progressive-approach-refugee-management\nYuval-Davis, N. (2006). Belonging and the politics of belonging. Patterns of Prejudice, 40(3), 197-214. doi:doi:10.1080/00313220600769331"}],"shard":{"index":24,"name":"turns-00024.parquet","rows":74005}}