Medical Education 1990, 24, 230-238
Academic, social and cultural factors influencing medical school grade performance S. F. ALFAYEZ, D . A. S T R A N D t S & J. D. CARLINES tDepartment of Medicine and Allied Sciences, Faculty of Medicine, King Abdulaziz University, Jeddah, fMedical Education Support Unit, Faculty of Medicine and Allied Sciences, King Abdulaziz University, Jeddah and #Division of Research in Medical Education, School of Medicine, University of Washington, Seattle Summary. Studies of medical student performance have focused on various factors, including premedical academics, maturity, familial background and support, and personal experiences withillness. Most studies have been conducted in countries with highly developed educational systems and similar cultural and social systems. It is not clear that these findings can be applied to developing countries, where the educational and cultural experiences may be very different, and where medical instruction is carried out in a non-native language. Information was obtained from a survey of 153 fifth- and sixth-year medical students at King Abdulaziz University in Saudi Arabia. The survey measured premedical educational, social and cultural experiences that might affect medical school performance. Men performed as well as women in the medical school despite heavy familial and social commitments. Women’s performance seems to be more influenced by changes in living environment. Achievement in premedical years was correlated positively with grade performance in medical school. Competence in the high-school English courses was related to medical school performance. Interest in the study of medicine prior to medical school was not related to performance. Other motivations, such as social gains, financial benefits or family wish, were related to lower performance. Current interest in clinical medicine correlated negatively with performance. Students motivated by the presence of chronic ill health in their families performed significantly
lower. Factors influencing medical school performance in developed countries had similar impact on medical students in a developing country. Social factors, unique to the country, also play a role in medical student performance. Key words: *students, medical; *educational status; *culture; *social environment; education, medical, undergraduate; family; Saudi Arabia
Introduction Performance of medical students attracts the attention of all those involved in medical education. Medical school selection committees, curriculum planners and instructional designers, as well as teachers, are concerned about student performance as it reflects on their various areas of interest. Grade performance serves, in many cases, as a basis for selection of medical students into medical education programmes. In some programmes, such as in Saudi Arabia, it is also the basis for the promotion and continuation of students through the medical school programme. Special interest has been given to factors which may be predictive of medical students’ performance. The goal is to use such information in selecting those students who are likely to perform well in the medical school. Included among these factors that have been investigated are the pre-academic achievement and medical college examination test (MCAT scores). The relationship of premedical grades to medical school grades has been reported in studies by Lipton et al. (1984) and Rippey et al. (1981). The studies of
Correspondence: Dr Saud F. Alfayez, Department of Medicine, King Abdulaziz University Hospital, PO Box 6615, Jeddah 21452, Saudi Arabia. 230
Factors inzuencing medical students’grades Murden et al. (1978) and Benor & Hobfoll(l984) show that premedical school scores are more related to performance in the basic science courses than to clinical course performance. However, Wingard & Williamson (1973), in an extensive review of the literature, concluded that both pre-academic and MCAT scores often reflect little more than ability to memorize isolated facts. Hunt et al. (1987) reported that performance o f medical residents who had difficulties as students was comparable to that oftheir peers except for two areas, namely ability to interact with patients and clinical decision-making. They were higher than their peers on dependability. Non-academic influences on performance have also been researched and reported by Rippey et al. (1981) and Rhoads et al. (1974). They reported that motivation appears to be an influencing factor in clinical performance as well as other non-academic factors such as maturity, rapport and non-academic achievement. Crimlisk & McManus (1987) hypothesized that personal or family experience with serious illness would be related to motivation in medical school, and stated that their inability to document a relationship was due to statistical technique rather than absence of a relationship. Hobfoll et al. (1982) and Huxham et al. (1980) in other studies concluded that personnlity factors showed no influence on clinical performance. Many of these studies have been conducted in medical schools in countries that have highly developed educational systems as well as similar cultural and social systems. However, during the past 25 years there have been massive transfers of medical education systems and technology to developing countries with different social and cultural systems. A question arises as to which factors influence medical school performance in developing countries. Are they the same as have been reported in the literature from the developed countries? Are high-school grades related to medical school grades in 6-year medical college programmes the same way as premedical grades for medical students from 4- or 8-year programmes? In most instances the English or French language is the main language used in medical education in developing countries. Saudi Arabia instructs medical students in English, even though it is not
23 1
the principal language of the students. Will students who achieved higher .grades in highschool English perform better than others in medical schools? Should admission standards be changed to better reflect the factors that influence student performance? Can curriculum development and instruction design be conducted similarly to medical education programmes in well-developed countries? Moreover, are family and social influences different for students in developing countries than in other parts of the world. Do these non-academic factors influence student’s medical school performance? Many social influences have been studied for students at the King Abdulaziz University College of Medicine (KAU-COM) in Jeddah. For example, about half of the students come from Jeddah and reside with their families. The rest are from outside Jeddah and live in housing on the university campus. Some students who live with their families report more outside commitments like driving for the family and helping their younger brothers and sisters in studying. O n the other hand, students who reside on campus are isolated from their family support systems and seem to suffer more from homesickness, an expected finding in a society where family ties are still quite strong. It would be important to determine if differences in performance exist between these two groups. Another factor is the difference in social commitments and distractions between men and women. It could be expected that women’s performance would be less affected by these social influences than the men, who have a greater demand on them in this area than the women. The purpose of this study is to examine some of these questions, as well as others, at King Abdulaziz University College of Medicine in Saudi Arabia.
Methods One hundred and forty-one fifth- and sixth-year medical students at King Abdulaziz University College of Medicine, Jeddah, Saudi Arabia were selected for this study. Data were collected for the 1986-87 academic year. The College of Medicine is a 6-year medical school programme
232
S. F . AIfayez et al.
followed by a 1-year internship and further residency training. A survey instrument was constructed that contained the following groups of questions: (1) Respondent data. The subjects were asked to identify their year in medical school, sex, age, marital status, nationality, place of residence, number of brothers and/or sisters and secondary school location. (2) Academic pegormance. Accumulated grade point averages (AGPA) were computed for student data for: total high-school years; highschool English courses and high-school science courses. Grades were categorized in five groups: excellent (90-100); very good (80-89); good (70-79); fair (60-69); and poor (0-59). The same approach was used to determine the students’ AGPA for: total medical school courses to date; basic science courses (years 1-3); and clinical science courses (years 4-6).
(3) Family intuence. Factors related to family influence on students’ selection and performance in medical school were computed as indices in four categories. ’Family educational level’ reflected the highest level of education attained both by parents and grandparents. ‘Family wish’ reflected the strength of the expressed desire of family members for the student to pursue a career in medicine. ‘Number of physicians in the family’ reflected the number of physicians in the student’s family and among relatives. ‘Chronic family illness’ reflected the strenth of influence of past family illnesses on the students’ decision to enter and pursue a career in medicine. (4) Interest as an influence in studying medicine and obtaining a high level of performance was also developed into an index. This index reflected the strength of interest motivating students in two categories: (a) interest prior to entering medical school; and (b) current interest in pursuing medical education. (5) Social intuences. An index was calculated to reflect social influences on the students’ choice and pursuit ofa medical career. Questions related to doctor income, doctor social status, societal expectations of students who receive the highest grades in high school, and the social demands and
responsibilities of family within the Saudi culture were included in this index. (6) Instructional inJuences. This index was broken down into three categories related to the self-reported influences in the class-room, library, laboratory, and clinic settings on students’ performance. ‘Language problems’ in class included the instructor’s ability to speak in English and the students’ ability to understand, speak and read English. ‘Teaching methods’ reflected the influence that the students reported feeling with regard to teaching methods, including schedule of classes, amount of homework, and quality of teaching. The ‘Limits of learning .resources’ index was calculated to reflect the students’ feeling about the strength o f influence that limitations in learning resources had on their performance. Questions focused on the amount of resources and opportunities to learning when attending clinics, emergency room work, number of patients in the hospital, and library and laboratory resources. The survey questionnaire was administered and explained to the students by one person during one of the periods scheduled for lecture near the end of the academic year in 1987. Those students who were absent or who were unable to complete the questionnaire in the allotted time were requested to take the instrument home and return it later. After a 2-week period, when students began their final examinations the instruments that had been returned were prepared for analysis. T w o hundred and thirty-seven instruments were passed out, and of these, 153 were returned. This resulted in an initial response rate of 65%. Twelve questionnaires were not usable due to incomplete data, leaving 141 usable survey instruments. This resulted in a final response rate of60%. Such a high return rate was not expected due to the sensitive nature of the family and social questions asked and the students’ preparation for the examinations. The total fifth- and sixth-year class included 133 men and 104 women. Men respondents to the survey were 43% of the total men in the classes while the women respondents were 81% of the women. This disparity in response was found to be statistically significant by x2 test (x2 = 146, P < 0.001). The distribution of academic performance of the
Factors inj7uencing medical students’grades respondents did not differ from that of the total class (x2 = 1.259, NS). Due to the differences between the rate of response for men and women students, further analyses were undertaken separately for each sex.
233
university hostel, while the remainder lived either at home with family or with relatives. Student’s t-test analysis between these two populations of students were performed and are reported in Table 2. No significant differences were noted for various variables studied, including family, interest, social and instructional influences. Moreover, medical school grades were not significantly different between the two groups. Men and women groups were then compared to determine whether there were any significant differences between them on the study variables. Student’s t-test group comparison yielded the results found in Table 3. There were significant differences between men and women groups on their total high-school AGPA (t = 3.39, P < 0.05). When the high-school English course (AGPAs) were compared there were also significant differences (P < 0.05) between the men (3.86) and women (4.30) means. There was also a significant difference between the men (4.45)and women (4.70)means for the high-school science courses. In all three cases the women students scored higher than the men on high-school
Results Demographic variables of respondents on sex, year in school, marital status, nationality, place of residence while in medical school, number of brothedsisters, and secondary school location are shown on Table 1. The greatest number of respondents were citizens of the Kingdom of Saudi Arabia who attended secondary school in the Kingdom. There was a fairly even distribution of fifth- (49%) and sixth- (51%) year students who responded to the survey. The average number of siblings was approximately six, which indicated that the students came from large families, which may have increased their family responsibilities, especially of those living at home. Thirty-eight per cent (38%) of the students lived away from home, predominantly at the
Table 1. Respondent characteristics: sex, year, school, marital status, nationality, place o f residence, average number o f brothers/sisters, and location o f secondary school Total students (n=141) n%
Male students (n=57) n%
Female students (n=84) n%
141
100
57
40
84
60
Year in school Fifth year Sixth year
69 72
49 51
31 26
22 18
38 46
27 33
Married students
24
17
11
8
13
9
134 7
95 5
57 0
40 0
77 7
55 5
Place o f residency Home or with relatives Away from home
87 54
62 38
35 22
25 15
53 32
37 23
Average number o f siblings
5.7
Location o f secondary school Saudi Arabia Outside Saudi Arabia
134 9
Sex
Nationality Saudi-Arabia Non-Saudi Arabia
5.9 95 5
57 0
5.6 40 0
77 7
55 5
* P < 0.05
Language difficulty Teaching methods Limited resources
Instructional injuences
Social injuences
Pre-college interest Current interest
Interest injluences
Education level No. MDs in family Chronic illness Wish to study medicine
Family inzuences
Total AGPA Basic science AGPA Clinical science AGPA
Medical school grades
Description
2.0 2.6 4.2
2.6
4.4
4.9 10.4 14.5
1.3 1.5
3.9 0.7 1.1 1.9
0.7 0.7 0.7
1.9 4.3
8.5 0-6 0.7 2.2
3.1 3.3 2.8
Live at home/with relatives (n = 87) Mean SD
4.9 10.0 13.7
3.6
43
1.9
7.2 0.5 0.9 2.2
3.0 3.1 2.7
2.1 3.3 5.4
2.6
1.2 1.5
2.9* 0.7 1.1 1.8
0.6 0.7 0.6
Live away from home hostel, etc. (n = 54) Mean SD
All respondents
5.1 10.2 14.4
4.9
14 4.0
8.3 0.6 0.7 1.7
3.1 3.3 2.7
1.9 2.6 4.7
2.7
1.3 1.5
3.9 0.8 1.o 1.8
0.7 0.7 0.7
Live at home/with relatives (n = 35) Mean SD
5.6 10.1 14.4
4.2
1.9 3.9
7.2 0.3 1.o 2.5
3.0 3.2 2.8
1.9 3.5 4.9
2.9
1.2 1.5
2.9 0.5 * 1.2 2.0
0.7 0.8 0.7
Live away from home hostel, etc. (n = 22) Mean SD
Men respondents
4.7 10.4 14.2
4.1
1.9 4.5
2.1 2.7 3.8
2.5
1.2 1.5
3.9 0.7 1.1 1.9
3.3 2.8 8.7 0.7 0.7 2.7
0.6 0.7 0.6
3.1
Live at homelwith relatives (n = 52) Mean SD
4.4 9.9 13.2
3.2
2.0 4.7
7.2 0.7 0.9 2.0
3.1 2.7
2.9
2.1 3.2 5.7
2.3
1.3 1.5
3.0 04 0.9 1.5
0.7 0.5
0.5
Live away from home hostel, etc. (n = 32) Mean SD
Women respondents
Table 2. Comparison of students' place of residence - living at home with family or relatives vs living away with study variables
L
4
n
$
F" ?I
Factors influencing medical students' grades
grades. This was not true for medical school grades as there were no significant differences calculated between men and women groups of the three medical school AGPAs. In two cases the men group mean was slightly higher than the women group. The only other study variable that yeilded a significant difference between the men and women groups was the current interest (or motivation) influence to study medicine. The women group index score suggested significantly greater current interest in pursuing their studies in medicine than the men groups (t = 2.85, P < 0.01). Pearson product moment correlations between the students' medical school performance (AGPAs) and study variables were calculated and reported in Table 4. Men and women students' combined high-school overall courses AGPA correlated positively with the basic sciences, clinical sciences and overall medical
235
school grade AGPAs. The correlation of highschool overall courses AGPA with basic science courses AGPA (0.31) was stronger than that with clinical science courses AGPA (0.15). Both correlations were statistically significant at P < 0.01 and P < 0.05, respectively. Men students had positive and moderate correlations between their total medical school AGPA and their high-school total courses AGPA (044), their English language courses AGPA (0.34) and their science courses AGPA (0.32). All these correlations were statistically significant ( P < 0.01). For women, the high-school scores correlated with basic science, clinical science and total medical school AGPAs. These correlations were less strong than those for men, 0.17, 0.12 and 0.18 respectively ( P < 0.05). Family influences showed variable effects on performance. A high family education level and presence of medical practitioners in the family yielded significant positive correlation with per-
Table 3. Comparison o f men students with women students on study variables Men Description
Women value
Mean
SD
Mean
SD
High-school grades Total AGPA English AGPA Science AGPA
4.58 3.86 4.45
0.53 0.9 1 0,68
4.84 4.30. 4.70
0.36 1.17 0.69
- 3.39"
Medical school grades Total AGPA Basic science AGPA Clinical science AGPA
3.08 3.28 2.73
0.67 0.73 0.74
3.06 3.22 2.76
0.6 1 0.69 0.61
0.15 0.49 - 0.27
Family inpuences Education level No.MDs in family Chronic illness Wish to study medicine
8.00 0.05 0.83 1.94
3.52 0.7 1 1.10 1.92
7.84 0.62 8.8 1 2.40
3.55 0.69 1.04 1.76
- 1.45
Interest injuerices Pre-college interest Current interest
1.84 4.52
1.27 0.99
1.97 5.03
1.25 0.96
- 2.85*
Social inpuences
4.62
2.74
3.88
2.41
1.66
5.26 10.43 14.58
1.94 2.73 4.43
4.66 10.36 14.06
2.06 2.82 4.49
1.76 0.15 0.68
Instructional influences Language difficulty Teaching methods Limited resources
* P < 0.05 ** P < 0.01
t
- 2.54* - 2.15*
0.25
- 0.98 0.07
- 0.62
238
S. F. A l f a y e z et al.
their studies. It may have been more difficult for them to attend out-patient clinics and the emergency room. Alternatively, the negative relation between interest and performance may have resulted from poorer performers providing misleading self-report answers in an attempt to improve their image. It is interesting to note that students who were motivated to join the medical school by the presence of chronic ill health in their families performed significantly lower than other students. This finding is difficult to explain. We can only speculate that this may have been due to the nature of that group whose decision-making seems to be strongly influenced by family and emotional factors. This might render the study of medicine even harder for them especially in subjects like anatomy. This study supports, in a different cultural context, findings reported in the West. In this developing country total high-school and English language scores were predictors of performance of medical students, especially in the basic sciences. The lack of strong predictors of academic performance in this study, particularly when compared to other studies, cannot be easily explained. The results here are similar to those found in another study of a Saudi Arabian medical school (El-Hazmi et al. 1987) and consequently may represent a factor(s) unique either to the medical schools and their grading practices o r to the students found in the Kingdom. Prior academic ability, tempered by English language proficiency, remains the most important predictor of performance in this medical school. While academic factors and social influences included here may be and probably are interrelated, this study represents an early attempt to explore the correlation of student performances in a developing medical school. Social factors showed some influence on performance in the medical school, albeit less strong than high-school scores. Social factors have been reported to influence subsequent career choice in
subspecialties of medicine. A study of this correlation is planned for the future. References Benor D.E. & Hobfoll S.E.(1984) Prediction ofclinical performance of medical students’ integrative approach to evaluation. Medical Education 18, 236-43. Crimlisk H.L. & McManus I.C. (1987) The effect of personal illness experience on career preference in medical students. Medical Education 21, 464-7. El-Hazmi M.A., Tekian A.S., El-Mahdy S. & Lambourne A. (1987) Performance of men and women medical students at King Saud University, Riyadh: a 10-year retrospective study at the College ofMedicineat King Saud University, between 1975 and 1985. Medical Education 21, 358-61. Hobfoll S.E.,Anson 0. & Antonovsky A. (1982) Personality factors as predictors of medical student performance. Medical Education 16,251-8. Hunt D.D., M.D., Scott C.S., Phil1ipsT.J.. Yergan A. & Greig L.M. (1987) Performance ofresidents who had academic difficulties in medical school. journal of Medical Education 62, 170-6. Huxam G.J., Lipton A. & Milton D.H. (1980) Achievement factors and personality in a cohort of medical students. Medical Education 14, 97-104. Lipton A., Huxham G.J. & Hamilton D. (1984) Predictors of success in a cohort of medical students. Medical Education 18, 203-10. Murden R., Galloway G.M., ReidJ.C. & Colwill J.M. (1978) Academic and personal predictors of clinical success in medical school. Journal of Medical Education 53, 71 1-9. Rhoads J.M., Gallemore J.L., Gianturco D.T. & Osterhout S., (1974) Motivation, medical school admissions, and student performance. journal OfMedical Education 49, 1119-27. Rippey R.M., Thal S . & Bongard S.J. (1981) A study of the University of Connecticut’s criteria for admission into medical school. Medical Education 15, 298-305. Wingard J.R. & Williamson J.W. (1973) Grades as predictors of physicians career performance: an evaluation literature review, Journal of Medical Educatioa 48, 312.
Received 9 February 1988; editorial comments to authors 1 3 October 1988; accepted f o r publication 27 January 1989
Academic, social and cultural factors influencing medical school grade performance S. F. ALFAYEZ, D . A. S T R A N D t S & J. D. CARLINES tDepartment of Medicine and Allied Sciences, Faculty of Medicine, King Abdulaziz University, Jeddah, fMedical Education Support Unit, Faculty of Medicine and Allied Sciences, King Abdulaziz University, Jeddah and #Division of Research in Medical Education, School of Medicine, University of Washington, Seattle Summary. Studies of medical student performance have focused on various factors, including premedical academics, maturity, familial background and support, and personal experiences withillness. Most studies have been conducted in countries with highly developed educational systems and similar cultural and social systems. It is not clear that these findings can be applied to developing countries, where the educational and cultural experiences may be very different, and where medical instruction is carried out in a non-native language. Information was obtained from a survey of 153 fifth- and sixth-year medical students at King Abdulaziz University in Saudi Arabia. The survey measured premedical educational, social and cultural experiences that might affect medical school performance. Men performed as well as women in the medical school despite heavy familial and social commitments. Women’s performance seems to be more influenced by changes in living environment. Achievement in premedical years was correlated positively with grade performance in medical school. Competence in the high-school English courses was related to medical school performance. Interest in the study of medicine prior to medical school was not related to performance. Other motivations, such as social gains, financial benefits or family wish, were related to lower performance. Current interest in clinical medicine correlated negatively with performance. Students motivated by the presence of chronic ill health in their families performed significantly
lower. Factors influencing medical school performance in developed countries had similar impact on medical students in a developing country. Social factors, unique to the country, also play a role in medical student performance. Key words: *students, medical; *educational status; *culture; *social environment; education, medical, undergraduate; family; Saudi Arabia
Introduction Performance of medical students attracts the attention of all those involved in medical education. Medical school selection committees, curriculum planners and instructional designers, as well as teachers, are concerned about student performance as it reflects on their various areas of interest. Grade performance serves, in many cases, as a basis for selection of medical students into medical education programmes. In some programmes, such as in Saudi Arabia, it is also the basis for the promotion and continuation of students through the medical school programme. Special interest has been given to factors which may be predictive of medical students’ performance. The goal is to use such information in selecting those students who are likely to perform well in the medical school. Included among these factors that have been investigated are the pre-academic achievement and medical college examination test (MCAT scores). The relationship of premedical grades to medical school grades has been reported in studies by Lipton et al. (1984) and Rippey et al. (1981). The studies of
Correspondence: Dr Saud F. Alfayez, Department of Medicine, King Abdulaziz University Hospital, PO Box 6615, Jeddah 21452, Saudi Arabia. 230
Factors inzuencing medical students’grades Murden et al. (1978) and Benor & Hobfoll(l984) show that premedical school scores are more related to performance in the basic science courses than to clinical course performance. However, Wingard & Williamson (1973), in an extensive review of the literature, concluded that both pre-academic and MCAT scores often reflect little more than ability to memorize isolated facts. Hunt et al. (1987) reported that performance o f medical residents who had difficulties as students was comparable to that oftheir peers except for two areas, namely ability to interact with patients and clinical decision-making. They were higher than their peers on dependability. Non-academic influences on performance have also been researched and reported by Rippey et al. (1981) and Rhoads et al. (1974). They reported that motivation appears to be an influencing factor in clinical performance as well as other non-academic factors such as maturity, rapport and non-academic achievement. Crimlisk & McManus (1987) hypothesized that personal or family experience with serious illness would be related to motivation in medical school, and stated that their inability to document a relationship was due to statistical technique rather than absence of a relationship. Hobfoll et al. (1982) and Huxham et al. (1980) in other studies concluded that personnlity factors showed no influence on clinical performance. Many of these studies have been conducted in medical schools in countries that have highly developed educational systems as well as similar cultural and social systems. However, during the past 25 years there have been massive transfers of medical education systems and technology to developing countries with different social and cultural systems. A question arises as to which factors influence medical school performance in developing countries. Are they the same as have been reported in the literature from the developed countries? Are high-school grades related to medical school grades in 6-year medical college programmes the same way as premedical grades for medical students from 4- or 8-year programmes? In most instances the English or French language is the main language used in medical education in developing countries. Saudi Arabia instructs medical students in English, even though it is not
23 1
the principal language of the students. Will students who achieved higher .grades in highschool English perform better than others in medical schools? Should admission standards be changed to better reflect the factors that influence student performance? Can curriculum development and instruction design be conducted similarly to medical education programmes in well-developed countries? Moreover, are family and social influences different for students in developing countries than in other parts of the world. Do these non-academic factors influence student’s medical school performance? Many social influences have been studied for students at the King Abdulaziz University College of Medicine (KAU-COM) in Jeddah. For example, about half of the students come from Jeddah and reside with their families. The rest are from outside Jeddah and live in housing on the university campus. Some students who live with their families report more outside commitments like driving for the family and helping their younger brothers and sisters in studying. O n the other hand, students who reside on campus are isolated from their family support systems and seem to suffer more from homesickness, an expected finding in a society where family ties are still quite strong. It would be important to determine if differences in performance exist between these two groups. Another factor is the difference in social commitments and distractions between men and women. It could be expected that women’s performance would be less affected by these social influences than the men, who have a greater demand on them in this area than the women. The purpose of this study is to examine some of these questions, as well as others, at King Abdulaziz University College of Medicine in Saudi Arabia.
Methods One hundred and forty-one fifth- and sixth-year medical students at King Abdulaziz University College of Medicine, Jeddah, Saudi Arabia were selected for this study. Data were collected for the 1986-87 academic year. The College of Medicine is a 6-year medical school programme
232
S. F . AIfayez et al.
followed by a 1-year internship and further residency training. A survey instrument was constructed that contained the following groups of questions: (1) Respondent data. The subjects were asked to identify their year in medical school, sex, age, marital status, nationality, place of residence, number of brothers and/or sisters and secondary school location. (2) Academic pegormance. Accumulated grade point averages (AGPA) were computed for student data for: total high-school years; highschool English courses and high-school science courses. Grades were categorized in five groups: excellent (90-100); very good (80-89); good (70-79); fair (60-69); and poor (0-59). The same approach was used to determine the students’ AGPA for: total medical school courses to date; basic science courses (years 1-3); and clinical science courses (years 4-6).
(3) Family intuence. Factors related to family influence on students’ selection and performance in medical school were computed as indices in four categories. ’Family educational level’ reflected the highest level of education attained both by parents and grandparents. ‘Family wish’ reflected the strength of the expressed desire of family members for the student to pursue a career in medicine. ‘Number of physicians in the family’ reflected the number of physicians in the student’s family and among relatives. ‘Chronic family illness’ reflected the strenth of influence of past family illnesses on the students’ decision to enter and pursue a career in medicine. (4) Interest as an influence in studying medicine and obtaining a high level of performance was also developed into an index. This index reflected the strength of interest motivating students in two categories: (a) interest prior to entering medical school; and (b) current interest in pursuing medical education. (5) Social intuences. An index was calculated to reflect social influences on the students’ choice and pursuit ofa medical career. Questions related to doctor income, doctor social status, societal expectations of students who receive the highest grades in high school, and the social demands and
responsibilities of family within the Saudi culture were included in this index. (6) Instructional inJuences. This index was broken down into three categories related to the self-reported influences in the class-room, library, laboratory, and clinic settings on students’ performance. ‘Language problems’ in class included the instructor’s ability to speak in English and the students’ ability to understand, speak and read English. ‘Teaching methods’ reflected the influence that the students reported feeling with regard to teaching methods, including schedule of classes, amount of homework, and quality of teaching. The ‘Limits of learning .resources’ index was calculated to reflect the students’ feeling about the strength o f influence that limitations in learning resources had on their performance. Questions focused on the amount of resources and opportunities to learning when attending clinics, emergency room work, number of patients in the hospital, and library and laboratory resources. The survey questionnaire was administered and explained to the students by one person during one of the periods scheduled for lecture near the end of the academic year in 1987. Those students who were absent or who were unable to complete the questionnaire in the allotted time were requested to take the instrument home and return it later. After a 2-week period, when students began their final examinations the instruments that had been returned were prepared for analysis. T w o hundred and thirty-seven instruments were passed out, and of these, 153 were returned. This resulted in an initial response rate of 65%. Twelve questionnaires were not usable due to incomplete data, leaving 141 usable survey instruments. This resulted in a final response rate of60%. Such a high return rate was not expected due to the sensitive nature of the family and social questions asked and the students’ preparation for the examinations. The total fifth- and sixth-year class included 133 men and 104 women. Men respondents to the survey were 43% of the total men in the classes while the women respondents were 81% of the women. This disparity in response was found to be statistically significant by x2 test (x2 = 146, P < 0.001). The distribution of academic performance of the
Factors inj7uencing medical students’grades respondents did not differ from that of the total class (x2 = 1.259, NS). Due to the differences between the rate of response for men and women students, further analyses were undertaken separately for each sex.
233
university hostel, while the remainder lived either at home with family or with relatives. Student’s t-test analysis between these two populations of students were performed and are reported in Table 2. No significant differences were noted for various variables studied, including family, interest, social and instructional influences. Moreover, medical school grades were not significantly different between the two groups. Men and women groups were then compared to determine whether there were any significant differences between them on the study variables. Student’s t-test group comparison yielded the results found in Table 3. There were significant differences between men and women groups on their total high-school AGPA (t = 3.39, P < 0.05). When the high-school English course (AGPAs) were compared there were also significant differences (P < 0.05) between the men (3.86) and women (4.30) means. There was also a significant difference between the men (4.45)and women (4.70)means for the high-school science courses. In all three cases the women students scored higher than the men on high-school
Results Demographic variables of respondents on sex, year in school, marital status, nationality, place of residence while in medical school, number of brothedsisters, and secondary school location are shown on Table 1. The greatest number of respondents were citizens of the Kingdom of Saudi Arabia who attended secondary school in the Kingdom. There was a fairly even distribution of fifth- (49%) and sixth- (51%) year students who responded to the survey. The average number of siblings was approximately six, which indicated that the students came from large families, which may have increased their family responsibilities, especially of those living at home. Thirty-eight per cent (38%) of the students lived away from home, predominantly at the
Table 1. Respondent characteristics: sex, year, school, marital status, nationality, place o f residence, average number o f brothers/sisters, and location o f secondary school Total students (n=141) n%
Male students (n=57) n%
Female students (n=84) n%
141
100
57
40
84
60
Year in school Fifth year Sixth year
69 72
49 51
31 26
22 18
38 46
27 33
Married students
24
17
11
8
13
9
134 7
95 5
57 0
40 0
77 7
55 5
Place o f residency Home or with relatives Away from home
87 54
62 38
35 22
25 15
53 32
37 23
Average number o f siblings
5.7
Location o f secondary school Saudi Arabia Outside Saudi Arabia
134 9
Sex
Nationality Saudi-Arabia Non-Saudi Arabia
5.9 95 5
57 0
5.6 40 0
77 7
55 5
* P < 0.05
Language difficulty Teaching methods Limited resources
Instructional injuences
Social injuences
Pre-college interest Current interest
Interest injluences
Education level No. MDs in family Chronic illness Wish to study medicine
Family inzuences
Total AGPA Basic science AGPA Clinical science AGPA
Medical school grades
Description
2.0 2.6 4.2
2.6
4.4
4.9 10.4 14.5
1.3 1.5
3.9 0.7 1.1 1.9
0.7 0.7 0.7
1.9 4.3
8.5 0-6 0.7 2.2
3.1 3.3 2.8
Live at home/with relatives (n = 87) Mean SD
4.9 10.0 13.7
3.6
43
1.9
7.2 0.5 0.9 2.2
3.0 3.1 2.7
2.1 3.3 5.4
2.6
1.2 1.5
2.9* 0.7 1.1 1.8
0.6 0.7 0.6
Live away from home hostel, etc. (n = 54) Mean SD
All respondents
5.1 10.2 14.4
4.9
14 4.0
8.3 0.6 0.7 1.7
3.1 3.3 2.7
1.9 2.6 4.7
2.7
1.3 1.5
3.9 0.8 1.o 1.8
0.7 0.7 0.7
Live at home/with relatives (n = 35) Mean SD
5.6 10.1 14.4
4.2
1.9 3.9
7.2 0.3 1.o 2.5
3.0 3.2 2.8
1.9 3.5 4.9
2.9
1.2 1.5
2.9 0.5 * 1.2 2.0
0.7 0.8 0.7
Live away from home hostel, etc. (n = 22) Mean SD
Men respondents
4.7 10.4 14.2
4.1
1.9 4.5
2.1 2.7 3.8
2.5
1.2 1.5
3.9 0.7 1.1 1.9
3.3 2.8 8.7 0.7 0.7 2.7
0.6 0.7 0.6
3.1
Live at homelwith relatives (n = 52) Mean SD
4.4 9.9 13.2
3.2
2.0 4.7
7.2 0.7 0.9 2.0
3.1 2.7
2.9
2.1 3.2 5.7
2.3
1.3 1.5
3.0 04 0.9 1.5
0.7 0.5
0.5
Live away from home hostel, etc. (n = 32) Mean SD
Women respondents
Table 2. Comparison of students' place of residence - living at home with family or relatives vs living away with study variables
L
4
n
$
F" ?I
Factors influencing medical students' grades
grades. This was not true for medical school grades as there were no significant differences calculated between men and women groups of the three medical school AGPAs. In two cases the men group mean was slightly higher than the women group. The only other study variable that yeilded a significant difference between the men and women groups was the current interest (or motivation) influence to study medicine. The women group index score suggested significantly greater current interest in pursuing their studies in medicine than the men groups (t = 2.85, P < 0.01). Pearson product moment correlations between the students' medical school performance (AGPAs) and study variables were calculated and reported in Table 4. Men and women students' combined high-school overall courses AGPA correlated positively with the basic sciences, clinical sciences and overall medical
235
school grade AGPAs. The correlation of highschool overall courses AGPA with basic science courses AGPA (0.31) was stronger than that with clinical science courses AGPA (0.15). Both correlations were statistically significant at P < 0.01 and P < 0.05, respectively. Men students had positive and moderate correlations between their total medical school AGPA and their high-school total courses AGPA (044), their English language courses AGPA (0.34) and their science courses AGPA (0.32). All these correlations were statistically significant ( P < 0.01). For women, the high-school scores correlated with basic science, clinical science and total medical school AGPAs. These correlations were less strong than those for men, 0.17, 0.12 and 0.18 respectively ( P < 0.05). Family influences showed variable effects on performance. A high family education level and presence of medical practitioners in the family yielded significant positive correlation with per-
Table 3. Comparison o f men students with women students on study variables Men Description
Women value
Mean
SD
Mean
SD
High-school grades Total AGPA English AGPA Science AGPA
4.58 3.86 4.45
0.53 0.9 1 0,68
4.84 4.30. 4.70
0.36 1.17 0.69
- 3.39"
Medical school grades Total AGPA Basic science AGPA Clinical science AGPA
3.08 3.28 2.73
0.67 0.73 0.74
3.06 3.22 2.76
0.6 1 0.69 0.61
0.15 0.49 - 0.27
Family inpuences Education level No.MDs in family Chronic illness Wish to study medicine
8.00 0.05 0.83 1.94
3.52 0.7 1 1.10 1.92
7.84 0.62 8.8 1 2.40
3.55 0.69 1.04 1.76
- 1.45
Interest injuerices Pre-college interest Current interest
1.84 4.52
1.27 0.99
1.97 5.03
1.25 0.96
- 2.85*
Social inpuences
4.62
2.74
3.88
2.41
1.66
5.26 10.43 14.58
1.94 2.73 4.43
4.66 10.36 14.06
2.06 2.82 4.49
1.76 0.15 0.68
Instructional influences Language difficulty Teaching methods Limited resources
* P < 0.05 ** P < 0.01
t
- 2.54* - 2.15*
0.25
- 0.98 0.07
- 0.62
238
S. F. A l f a y e z et al.
their studies. It may have been more difficult for them to attend out-patient clinics and the emergency room. Alternatively, the negative relation between interest and performance may have resulted from poorer performers providing misleading self-report answers in an attempt to improve their image. It is interesting to note that students who were motivated to join the medical school by the presence of chronic ill health in their families performed significantly lower than other students. This finding is difficult to explain. We can only speculate that this may have been due to the nature of that group whose decision-making seems to be strongly influenced by family and emotional factors. This might render the study of medicine even harder for them especially in subjects like anatomy. This study supports, in a different cultural context, findings reported in the West. In this developing country total high-school and English language scores were predictors of performance of medical students, especially in the basic sciences. The lack of strong predictors of academic performance in this study, particularly when compared to other studies, cannot be easily explained. The results here are similar to those found in another study of a Saudi Arabian medical school (El-Hazmi et al. 1987) and consequently may represent a factor(s) unique either to the medical schools and their grading practices o r to the students found in the Kingdom. Prior academic ability, tempered by English language proficiency, remains the most important predictor of performance in this medical school. While academic factors and social influences included here may be and probably are interrelated, this study represents an early attempt to explore the correlation of student performances in a developing medical school. Social factors showed some influence on performance in the medical school, albeit less strong than high-school scores. Social factors have been reported to influence subsequent career choice in
subspecialties of medicine. A study of this correlation is planned for the future. References Benor D.E. & Hobfoll S.E.(1984) Prediction ofclinical performance of medical students’ integrative approach to evaluation. Medical Education 18, 236-43. Crimlisk H.L. & McManus I.C. (1987) The effect of personal illness experience on career preference in medical students. Medical Education 21, 464-7. El-Hazmi M.A., Tekian A.S., El-Mahdy S. & Lambourne A. (1987) Performance of men and women medical students at King Saud University, Riyadh: a 10-year retrospective study at the College ofMedicineat King Saud University, between 1975 and 1985. Medical Education 21, 358-61. Hobfoll S.E.,Anson 0. & Antonovsky A. (1982) Personality factors as predictors of medical student performance. Medical Education 16,251-8. Hunt D.D., M.D., Scott C.S., Phil1ipsT.J.. Yergan A. & Greig L.M. (1987) Performance ofresidents who had academic difficulties in medical school. journal of Medical Education 62, 170-6. Huxam G.J., Lipton A. & Milton D.H. (1980) Achievement factors and personality in a cohort of medical students. Medical Education 14, 97-104. Lipton A., Huxham G.J. & Hamilton D. (1984) Predictors of success in a cohort of medical students. Medical Education 18, 203-10. Murden R., Galloway G.M., ReidJ.C. & Colwill J.M. (1978) Academic and personal predictors of clinical success in medical school. Journal of Medical Education 53, 71 1-9. Rhoads J.M., Gallemore J.L., Gianturco D.T. & Osterhout S., (1974) Motivation, medical school admissions, and student performance. journal OfMedical Education 49, 1119-27. Rippey R.M., Thal S . & Bongard S.J. (1981) A study of the University of Connecticut’s criteria for admission into medical school. Medical Education 15, 298-305. Wingard J.R. & Williamson J.W. (1973) Grades as predictors of physicians career performance: an evaluation literature review, Journal of Medical Educatioa 48, 312.
Received 9 February 1988; editorial comments to authors 1 3 October 1988; accepted f o r publication 27 January 1989
