Introduction

The educational environment consists of, among other factors, a documented curriculum and a perception of teacher–student interaction both in and out of class (1). The anticipation and experiences of students concerning these factors can either motivate or demotivate them in their learning approaches and learning style, crucial to these being how the learner is engaged (2).
The learning environment as perceived by students determines their behavior and aspirations (3). The learning  environment is one of the factors affecting quality that can be measured and actions taken to improve, correct, or change the environment in order to improve the quality of education (3,4). The Dundee Ready Education Environment Measure (DREEM) is a culturally non-specific tool that has been applied in many cultures and languages. It has been validated to be good for measuring the learning environment among medical students (5,6). It has also been used to determine the weaknesses or strengths of an institution’s learning environments, make comparative analyses of students’ perceptions within and between different cohorts, and can identify academic achievers (7).
To the best of our knowledge no study using the DREEM tool or any other tool to determine students’ perceptions of the learning environment has previously been carried out in Kenya or at the University of Nairobi; thus, this study was undertaken.
                                                       

Materials and Methods

This was a cross-sectional survey of 3rd to 6th year medical students at the School of Medicine of the University of Nairobi. Ethical approval was obtained from Kenyatta National Hospital–University of Nairobi Ethics and Research Committee (P55/01/2019). We used the DREEM tool to gauge students’ perceptions of the educational environment.
DREEM contains 50 statements on a range of topics directly relevant to the medical educational environment, scored on a 5-point Likert scale ranging from 0 to 4 (4: strongly agree–SA; 3: agree; 2: have no idea; 1: disagree; 0: strongly disagree–SD). The 50 items have a maximum score of 200. However, 9of the 50 (8, 12, 15, 16, 21, 23, 34, 39, 45) were given a reverse score, i.e., 0=SA and 4=SD, so that a higher score would indicate more positive responses. The inventory encompasses five subscales: (1) students’ perceptions of learning (SPL)–12 items; (2) students’ perceptions of teachers (SPT)–11 items; (3) students’ academic self-perception (SASP)–8items; (4) students’ perceptions of atmosphere (SPA)–12 items; and (5) students’ social self- perceptions (SSSP)–7items.
Students who agreed to participate consented in writing. The tool was completed during class after an explanation from one of the research assistants and returned at the end of class. Data collection for DREEM consisted of paper-based demographics that included year of study and gender.
Data were analyzed using SPSS version 19 (IBM SPSS Statistics for Windows, Version 19.0. Armonk, NY: IBM Corp). Analysis was on the overall score out of 200, subscale score (SPL–48, SPT–44, SASP–32, SPA–48, SSSP–28) and individual score (0–4). Overall score of 0–50 is interpreted as extremely poor, 51–100 indicates plenty of problems, 101–150 is more positive than negative, and 151–200 as excellent.
Individual items with a mean score ≥3.5 are particularly strong areas, items with a mean score ≤2.0 need attention, and items with mean scores between 2 and 3 are areas of the educational environment that could be improved. Cronbach’s alpha was calculated for internal validity of the DREEM tool within our context. The Kaiser–Meyer–Olkin test was performed for sampling adequacy. Comparison of mean used Student’s t-test and analysis of variance (ANOVA). Statistical significance was taken as p<0.05.
                                          

Results

We obtained a 77.4% response rate (619/800). The highest number of responses (32.1%) was obtained from 4th year students. The gender response rates were: female 50.2% and male 49.8% (Table 1).
Cronbach’s α was 0.882 while the Kaiser–Meyer–Olkin test was 0.904 (p<0.001), indicating sample size was adequate and had a high internal validity. The mean score was 93.3/200 (46.7%), implying that generally there are numerous problems with how our students perceive their educational experience.
On the subscale scores, only year V seems to have overall and specific areas such as SPL, SPT, and SSAP that appear to be moving towards a positive experience. In addition, year III has positives for SPT and year VI has positives for SSAP (Figure 1). The student perception on the learning subscale is an area that requires attention as the only statement that scored above 2 was “I am clear about the learning objectives of the course” (Figure 2).
ANOVA revealed significant differences in the scores across the clinical years in all questions except 47 (p=0.772) and 48 (p=0.072), all other questions had p less than 0.05. The differences in gender were significant only for questions 13 (p=0.006), 22(p=0.022), 44 (p=0.009), and 47(p=0.012).

​

Students’ perceptions of teachers
Across this subscale, only statement 2 had a mean above 3.0. Areas that require action include every item statement where the score was less than 2 (Figure 3). A one-way ANOVA showed mean for year IV was significantly different from other class years in all the areas (p as follows: questions 2=0.004, 6=0.014, 8=<0.001, 9 =<0.001, 18=0.012, 19=<0.001, 32=0.014, 37=0.040, 39=<0.001, 40=<0.001, and 50=0.001). The score by females was significantly different from that of males on questions 19 and 32 with p of 0.007 and 0.014, respectively.

 

Students’ academic self-perception
Most areas of academic self-perception need attention except two areas: statements 10 and 45 (Figure 4). A one-way between-subjects ANOVA compared the effect of classes on their academic self-perception for years III, IV, V and VI. The higher the student’s year the more likely they were to agree with item 5, p=0.027, while they were more likely to disagree with items 10, 21, 26, 27, and 41 with p=<0.001 for all items. Female students disagreed with items 5, 10, 21, and 27 with p=0.038, 0.005, <0.001 and <0.001, respectively. 

 

Students’ perceptions of atmosphere
Items 23,30, 33, 34 and 36 scored between 2-3 and therefore could be improved (and the rest need attention because they scored below 2 (Figure 5). Using ANOVA there was significant difference in means for year class with year IV disagreeing with statements 11, 12, 23, 30, 34, 35, 42, 43, and 49 and agreeing with statement 17, p values of <0.001, <0.001, 0.041, <0.001, 0.008, 0.003, 0.001, 0.011, <0.001, and 0.002 respectively. The females also agreed with statement 17(p=0.001), and disagreed with statements 11, 12, 42, and 49 with p=0.005, <0.001, 0.020, and 0.001, respectively.

 

Students’ social self-perception
The worst area is that of support for students, with a mean of less than 1 (Table 2). Using ANOVA, females agreed with item 4, p=0.006, and disagreed significantly with items 14, p=0.011, and 46, p<0.001. All classes disagreed, although more of year IV disagreed significantly with statements 3 ,14 ,29, and 46, than agreed with statement 4 with p=<0.001, 0.008, <0.001, <0.001, and 0.009 respectively.

​

Click to view Table 1

Click to view Table 2

​

Click to view Figure 1

Click to view Figure 2

Click to view Figure  3

Click to view Figure  4

Click to view Figure  5

​

Discussion

The DREEM instrument was developed and validated for use in any culture and country (6). It has been used to measure the educational environment, identify problem areas in an education program, get a baseline before instigating curriculum reform, determine student reaction to curriculum reform, and as a tool for monitoring improvement (8,9). In this study, it was used for assessing students’ perceptions of the learning environment and possible problem areas in the education program.
The overall score was 93.3/200, which implies perception of a very poor learning environment with plenty of problems. Similar findings have been reported by various studies (10–13),mostly in the Asian and Arab worlds. Factors associated with low scores in these studies were the use of a traditional curriculum instead of a modern curriculum that includes problem-based learning or system-based learning (11,14). The more senior classes were more likely to give lower scores (10,13,15) in environments where teachers are perceived to be authoritarian (10). Most surveys usually get an average of between 101 and150, meaning more positive than negative responses, as was described in a systematic review of DREEM studies by Chan et al. (16). Only a few studies have scored higher than 150 according to Chan et al.’s systematic review, and those that scored high had small to moderate size classes (17) and those that had placement in general practice or community hospitals (18).
The medical school is an institution whose principal inhabitants are its students; its raison d’êtreis their education, training, and welfare (19). It is for this reason that student’s perception of their learning environment is an important factor. A deeper look at this complex environment, both in the subscale and individual items, reveals the depth of the problems.
The SPL domain had only one statement (I am clear about the learning objectives of the course) scoring above 2. The only year that scored most items above 2 was year V. The reason could be that students are divided into smaller groups that deal with a particular discipline before moving to the next. They may therefore perceive this arrangement as a more friendly micro-environment created in this space. The most affected are the 4th years who seem to be finding the clinical years difficult and the teachers hostile to them. This result is similar to the study by Hasan and Gupta in Iran and Roff et al. in comparing Nigeria and Nepal where most scores were low in earlier years, then increased, and then declined again (10,20). Emphasis is on factual learning, superfluous teacher-centered learning where lecturers just read through the slides, as well as an overemphasis on exam scores. This is similar to studies done in Iran (21) and Sri Lanka (22) and in contrast to the study by Abraham et al. in India (23). Another study has shown that students in traditional medical curricula often perceived learning as being too teacher centered, dogmatic, and with an over-emphasis on “rote memorization” (24).
The SPT domain had 21.73/44 (49.3%) which is above all other domains in this study, with just over half of the items being above 2 and one above 3. The students acknowledge that their teachers are knowledgeable but may lack the critical skills in how to deliver the subjects. They also criticize teachers as being unavailable, irritable, authoritarian, and having difficulty in interacting with patients and giving feedback to students. Teachers being irritable, sarcastic, and authoritarian has long been noted by Abraham et al. in India (23) and Lokuhetty et al. in Sri-Lanka (22) as well as Hasan and Gupta in Iran (10). This has been postulated to be due to pressure of work and increased work load which might be eased by even distribution of work (10).
In the SPA domain, most items, 7 out of 12, were below 2, indicating serious problems. They include lack of following the schedule, tension in classes and in wards due to the interaction of the teachers with the students expressing social discomfort, an inability to clarify doubts in class due to being chastised by teachers, cheating by other students, and a lack of interpersonal cohesiveness. These factors further diminish confidence among students in the environment and coupled with heterogeneity (male–female, Kenyan–non-Kenyan, age) of the student population, bring about inadequate bonding patterns. The diminishing of student confidence is further reduced by hectic student schedules and authoritarian or uninvolved teaching styles. Hasan and Gupta as well as Hassan et al. found similar issues in Iran and Sudan respectively (10,24). These issues should be explored in remedial measures because they lead to failure due to factors other than the academic ability of the students.
In the SASP domain, 6th year students seem more confident than the rest of the students. Increased confidence among seniors has been noticed in the studies in Asia (21,22).
One of the poorly scored domains was SSP, mainly because of the lack of social support during stressful periods, as has been shown in the SPA domain. This could arise due to a lack of teacher interaction and involvement in student life. The involvement may be due to the teacher’s workload, among other things. So strong was this perceived lack of support and the teacher’s role in increasing stress that some felt the School of Medicine participates in destroying and not building the life and career of students. The stress in medical school may be attributed to exam anxiety, lack of leisure time, and inadequate resources as shown by studies from Australia (25) and Europe (26).
Limitations of this study include: it is a single-center study, timing of the study in that it was given 4 weeks before exams, and the increased stress associated with exams could have led to more complaints and participant bias.
                                            

Conclusion

This study shows that medical students in the University of Nairobi are critical of the general climate of the school and of the quality of teaching, especially in areas regarding class participation and provision of clear learning objectives. The students also perceive their courses as being tutor centered, having an unbalanced curriculum, having difficulty with time-tabling of teaching, having knowledgeable teachers who do not provide feedback in an appropriate manner, and the school being overly concerned with the marks obtained and not the clinical skills learned. These findings should encourage reflection by medical schools in Kenya to relook at the learning environment of medical education so as to improve it.

                                                                  

References

1. Roff S, Mcaleer S. What is educational climate? Med Teach. 2001;23(4):333–334. 
2. Hutchinson L. Educational environment. BMJ. 2003 ;326(7393) :810–812. 
3. Plucker JA. The relationship between school climate conditions and student aspirations. J Educ Res. 1998;91(4):240–246. 
4. Pimparyon SR, Caleer SM, Pemba S. Educational environment, student approaches to learning and academic achievement in a Thai nursing school. Med Teach. 2000;22(4):359–364. 
5. Kohli V, Dhaliwal U. Medical students' perception of the educational environment in a medical college in India: a cross-sectional study using the Dundee Ready Education Environment questionnaire. J Educ Eval Health Prof. 2013; 10:5. 
6. Roff S, McAleer S, Harden RM, et al. Development and validation of the Dundee Ready Education Environment Measure (DREEM), Med Teach. 1997;19(4):295–299
7. Mayya S, Roff S. Students’ perceptions of educational environment: a comparison of academic achievers and under-achievers at Kasturba medical college, India. Educ Health. 2004;17(3):280–291. 
8. Till H. Climate studies: can students’ perceptions of the ideal educational environment be of use for institutional planning and resource utilization? Med Teach. 2005;27(4):332–337. 
9. Edgren G, Haffling A-C, Jakobsson U, et al. Comparing the educational environment (as measured by DREEM) at two different stages of curriculum reform. Med Teach. 2010;32(6): e233–238. 
10. Hasan T, Gupta P. Assessing the learning environment at Jazan medical school of Saudi Arabia. Med Teach. 2013;35(Suppl. 1): S90–96. 
11. AlFaris EA, Naeem N, Irfan F, et al. Student-centered curricular elements are associated with a healthier educational environment and lower depressive symptoms in medical students. BMC Med Educ. 2014; 14:192. 
12. Koohpayehzadeh J, Hashemi A, Soltani Arabshahi K, et al. Assessing validity and reliability of Dundee ready educational environment measure (DREEM) in Iran. Med J Islam Repub Iran. 2014; 28:60. 
13. Belayachi J, Razine R, Boufars A, et al. Moroccan medical students' perceptions of their educational environment. J Educ Eval Health Prof. 2015; 12:47. 
14. Kim H, Jeong H, Jeon P, et al. Perception study of traditional Korean medical students on the medical education using the Dundee ready educational environment measure. Evidence-Based Complementary and Alternative Medicine. 2016; 2016:6042967.
15. Andalib MM, Malekzadeh MM, Agharahimi Z, et al. Evaluation of educational environment for medical students of a tertiary pediatric hospital in Tehran, using DREEM questionnaire. Iran J Pediatr. 2015;25(5): e2362.   
16. Chan CYW, Sum MY, Tan GMY, et al. Adoption and correlates of the Dundee Ready Educational Environment Measure (DREEM) in the evaluation of undergraduate learning environments–a systematic review. Med Teach. 2018;40(12):1240–1247.  
17. Bennett D, Kelly M, O’Flynn S. Are the bigger hospitals better: DREEM on? Ir J Med Sci. 2010; 179(4):515–219. 
18. Tackett S, Shochet R, Shilkofski NA, et al. Learning environment assessments of a single curriculum being taught at two medical schools 10,000 miles apart. BMC Med Educ. 2015; 15:105. 
19. Genn JM. AMEE Medical Education Guide No. 23 (Part 1): Curriculum, environment, climate, quality and change in medical education-a unifying perspective. Med Teach. 2001;23(4):337–344. 
20. Roff S, McAleer S, Ifere OS, et al. A global diagnostic tool for measuring educational environment: comparing Nigeria and Nepal. Med Teach. 2001;23(4):378–382. 
21. Taheri M. Students’ perceptions of learning environment in Guilan University of Medical Sciences. J Med Educ. 2009;13(4):126–133
22. Lokuhetty MD, Warnakulasuriya SP, Perera RI, et al. Students’ perceptions of learning environment in an Indian medical school. BMC Med Educ. 2008; 8(1): 20. 
23. Al-Hazimi A, Zaini R, Al-Hyiani A, et al. Educational environment in traditional and innovative medical schools: a study in four undergraduate medical schools. Educ Health. 2004;17(2):192–203. 
24. Hassan M, Abdullah DI. Measuring the medical educational environment at Alzaiem Alazhari University. Khartoum Med J. 2012;3(3). 
25. Sanders AE, Lushington K. Sources of stress for Australian dental students. J Dent Educ. 1999; 63(9):688–697. 
26. Jönsson M, Ojehagen A. Medical students experience more stress compared with other students. Lakartidningen. 2006;103(11):840.