Using an iterative mixed-methods research design to investigate schools facing exceptionally challenging circumstances within Trinidad and Tobago

ABSTRACT

In this study, methodology and theory were linked to explicate the nature of education practice within schools facing exceptionally challenging circumstances (SFECC) in Trinidad and Tobago. The research design was an iterative quan>QUAL-quan>qual multi-method research programme, consisting of 3 independent projects linked together by overall purpose. Although large-scale quantitative data were first used to identify and describe SFECC, the overall programme was driven by an inductive theoretical thrust, with findings from the multiple-site case studies of Projects 2 and 3 used to construct substantive theory on SFECC. The integrated findings point to the pervasive role of beyond-school factors, which were magnified and intertwined with between- and within-school factors. SFECC in this study found great difficulty transcending states of chronic low performance. We hypothesized on the need for transformative school leadership, but in this context, greater external support by stakeholder agencies and broad place-based strategies appeared essential for turnaround.

KEYWORDS:

• Schools facing exceptionally challenging circumstances
• high-poverty schools
• school effectiveness
• mixed-methods research
• Caribbean region
• Trinidad and Tobago

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. National tests are large-scale assessments used for monitoring. Mathematics and English Language Arts assessments are administered at Standards 1 and 3.

2. This measure varied from 0% to 155%. Figures above 100% are due to a policy of supplying additional lunches to all rural schools and Tobago. Notably in this regard, many rural schools are very small and located in isolated areas. Several rural schools have populations of fewer than 20 pupils.

3. Morse and Niehaus (2009) use the term core project to describe the project that is given emphasis in a qualitatively driven, multiple-method research programme.

4. This excludes the private primary schools (N = 62).

5. Stake (2006) suggests that the number of cases be 3 to 10.

6. These criteria were based on the findings from the inferential and multilevel analysis conducted in Project 1 and expressed views of the Division of Educational Research and Evaluation (DERE) of the Trinidad and Tobago Ministry of Education. Multilevel modelling suggested that geographic location might be important for SFECC school performance. The DERE was keen to understand how single sex and coeducational arrangements helped or hindered school performance in these contexts.

7. These classes are administered the National Tests (Standards 1 and 3) and Public Examinations (Standard 5).

8. Data on the school-level variables were obtained from the annual form submitted by all schools to the Division of Planning, Trinidad and Tobago Ministry of Education.

9. The Instructional Practices Inventory reports six categories: (1) student disengagement, (2) student work with teacher not engaged, (3) student work with teacher engaged, (4) teacher-led instruction, (5) student verbal learning conversations, and (6) student active engaged learning.

10. All teachers of target classes completed questionnaires on collective teacher efficacy and group organizational citizenship behaviour. However, data from these survey instruments are not reported in this study.

11. In this paper, we report only on the interview data with the principal.

12. The ICCs were as follows: for SFECC schools – Lang 1 (.277), Lang 3 (.193), Math 1 (.225), Math 3 (.171); for Rest of Schools – Lang 1 (.256), Lang 3 (.213), Math 1 (.198), Math 3 (.227).

13. For management, the schools were coded as (1) Government and (2) Government-assisted (managed by a denominational board).

14. Geographic location was classified as (1) urban, (2) mixed urban, (3) mixed rural, and (4) rural.

15. Multigrade status refers to the existence of multiple-year groups within one or more classrooms of the school. Schools were coded as (1) full multigrade, (2) partial multigrade, and (3) no multigrade.

16. In this study, modal profiles are reported. A modal profile is a qualitative categorization of a group based on the most frequently occurring attribute/s (Teddlie & Tashakkori, 2009).

17. According to Onwuegbuzie et al. (2009), the nature of generalization is a fundamental principle guiding data analysis.

18. Two SFECC schools and 16 of rest of schools were omitted from the HLM analysis because of missing data.

19. The coding used for geographic location was rural (4), mixed rural (3), mixed urban (2), and urban (1).

20. There were significant numbers of students in both Schools A and B from an adjacent community, which reported one of the lowest BNIs in the country.

21. The majority of observations recorded identified Categories 1 to 4 of the Instructional Practices Inventory (Categories 5 and 6 measure active and engaged learning).

22. Modal responses for No: School A 20 out of 41; School B 16 out of 42; School C 14 out of 14.

23. In analysis of the Progress in International Reading Literacy Study (PIRLS) 2006 data, 100 and more books is placed in the category labelled as “more resources” (Mullis, Martin, Kennedy, & Foy, 2007)

24. Parents responded to four categories: Never, Once, Twice, and More than three times.

25. See Figure 4 in De Lisle (2011) for photographic evidence.

26. See Figure 4 in De Lisle (2011).

Additional information

Funding

This work was supported by the The UWI-Trinidad and Tobago Research and Development Impact Fund.

Notes on contributors

Jerome De Lisle

Jerome De Lisle is Senior Lecturer in Educational Administration in the School of Education, The University of the West Indies, St. Augustine. His main research areas are in educational assessment, data use, and system reform. He has worked extensively as a consultant in the areas of educational evaluation and system reform with the Trinidad and Tobago Ministry of Education and is Principal Investigator in the local research projects: Schools Facing Exceptionally Challenging Circumstances and Schools on the Rise.

Krishna Seunarinesingh

Krishna Seunarinesingh is Lecturer, Teaching of English, at the School of Education, The University of the West Indies, Trinidad and Tobago. His research interests are related to developing English teachers’ pedagogical content knowledge, especially as it relates to the teaching of writing, and the use of instructional scaffolding.

Rhoda Mohammed

Rhoda Mohammed is an Educational Testing Officer with the Trinidad and Tobago Ministry of Education. She is currently a Postgraduate Student at the School of Education, The University of the West Indies, St Augustine. She is completing a PhD in Education, specializing in data-driven decision making in education.

Rinnelle Lee-Piggott

Rinnelle Lee-Piggott recently joined the University of the West Indies as a lecturer in Educational Administration, prior to which she functioned as a teacher leader at the primary school level in Trinidad. She holds a PhD in Education and an MA in Educational Leadership and Management awarded by the University of Nottingham and a BEd in Primary Education conferred by the University of the West Indies. Her research interests are in schools facing challenging circumstances, beginning principal leadership and development, teacher leadership, school culture, and school change.