

ORIGINAL ARTICLE 

Year : 2021  Volume
: 9
 Issue : 2  Page : 121125 

Learning statistics: Interprofessional survey of attitudes toward statistics using SATS36
Htoo Htoo Kyaw Soe^{1}, Sujata Khobragade^{1}, Htay Lwin^{1}, Mila Nu Nu Htay^{1}, Nan Nitra Than^{1}, Khine Lynn Phyu^{2}, Adinegara Lutfi Abas^{1}
^{1} Department of Community Medicine, MelakaManipal Medical College, Melaka, Malaysia ^{2} Department of Paediatrics, MelakaManipal Medical College, Melaka, Malaysia
Date of Submission  24Dec2020 
Date of Decision  07Jul2021 
Date of Acceptance  08Jul2021 
Date of Web Publication  30Nov2021 
Correspondence Address: Htoo Htoo Kyaw Soe MelakaManipal Medical College, Jalan Batu Hampar, Bukit Baru, 75150 Melaka Malaysia
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/dmr.dmr_68_20
Context: In medical education, statistics is part of the core training to enable the students to plan, design, analyze, and interpret the experimental data. It is believed that attitudes toward statistics play an important role in learning statistics. Aims: We conducted this study to assess the attitudes toward statistics among undergraduate medical and dentistry students. Settings and Design: This crosssectional study was done among 3^{rd}year medical students and 4^{th}year dentistry students in a private medical college in Malaysia. Subjects and Methods: We employed purposive sampling and invited the 3^{rd}year medical students and the 4^{th}year dentistry students before the commencement of the biostatistics course. A total of 206 students participated in this study. We utilized SATS36 scale which consisted of 36 items that were divided into six subscales such as affect, cognitive competence, value, difficulty, interest, and effort. Statistical Analysis Used: Data were analyzed using descriptive statistics, independent ttest, and Spearman's rho correlation. Results: The students had positive attitudes toward statistics in most of the domains of SATS36 except difficulty. There were no significant differences of attitudes toward statistics between male and female students. However, male students had a significantly higher mean score in the interest domain (mean difference 0.40 [95% confidence interval 0.07, 0.74]). Conclusions: The undergraduate medical and dentistry students had positive attitudes toward statistics, but the students found that statistics is a difficult subject. Hence, the instructors need to understand the student's attitudes and create effective learning strategies which not only provide knowledge and skills but also change student's attitudes toward the desired direction.
Keywords: Attitudes, dentistry, medical, statistics, students
How to cite this article: Kyaw Soe HH, Khobragade S, Lwin H, Nu Htay MN, Than NN, Phyu KL, Abas AL. Learning statistics: Interprofessional survey of attitudes toward statistics using SATS36. Dent Med Res 2021;9:1215 
How to cite this URL: Kyaw Soe HH, Khobragade S, Lwin H, Nu Htay MN, Than NN, Phyu KL, Abas AL. Learning statistics: Interprofessional survey of attitudes toward statistics using SATS36. Dent Med Res [serial online] 2021 [cited 2022 Jan 23];9:1215. Available from: https://www.dmrjournal.org/text.asp?2021/9/2/121/331402 
Introduction   
Statistics are employed in many fields such as business, education, psychology, agriculture, and economics.^{[1]} Colleges, tertiary, and higher educational institutions include some level of statistics in their curriculum though the level of necessities varies from major to major.^{[2],[3]} Medical statistics is a branch of statistics in which statistical theories are applied in medical and biological research. Nowadays, statistics are increasingly utilized in medical practice, especially when making a diagnosis or deciding appropriate treatment.^{[4]} In medical education, statistics is part of the core training to enable the students to plan, design, analyze, and interpret the experimental data.^{[5]}
Although sound statistical knowledge is fundamental to understanding and analyzing medical research, it is considered to be a course that is difficult to teach and learn.^{[3],[5],[6],[7]} Numerous statistics teachers as well as the students believe that attitudes toward statistics play an important role in learning statistics.^{[8]} The student's attitudes not only influence the classroom environment but also his/her attitudes affect their course completion and achievement and ability to apply the knowledge of statistics outside the classroom and in everyday life.^{[3],[8],[9]} It has been recognized that teaching and learning statistics involve dealing with the student's interest, anxieties, and negative attitudes toward the subject.^{[3],[8]}
In our college, students are having a compulsory research methodology course in the 3^{rd} year of the Bachelor of Medicine and Bachelor of Surgery (MBBS) program and the 4^{th} year of the Bachelor of Dental Surgery (BDS) program. The aim of this course is to introduce principles of scientific research and biostatistics using various problemsolving exercises and to provide the skills that can effectively contribute to various institutional research projects. In this course, students are required to learn basic biostatistics including descriptive statistics and hypothesis testing.^{[10]} Not only to improve student's learning process and to achieve successful course completion but also to develop positive attitudes toward statistics, it is important to understand the student's feelings concerning statistics, their value, and interest before the course. Therefore, we conducted this crosssectional study in our college to assess the attitudes toward statistics among undergraduate medical and dentistry students before the commencement of the biostatistics module and to determine the relationship between career value using statistics and student's attitudes.
Subjects and Methods   
This crosssectional study was done among the 3^{rd}year medical students and the 4^{th}year dentistry students from April to December 2018. There were 280 students attending the 3^{rd} year of MBBS program and 73 students attending the 4^{th} year of BDS program. The sample size was calculated using the formula for estimating the finite population proportion with the margin of error 5%, the assumption of 95% confidence level,^{[11]} and 71.11% of students having positive attitudes toward statistics.^{[3]} The minimum sample size required was 160. We included a nonresponse of 20% and the final sample size was 200. We employed purposive sampling and invited the 3^{rd}year medical students who were posted in community medicine from April to December 2018 and the 4^{th}year dentistry students (from November to December 2018) before the commencement of the biostatistics course. Students were explained that participation was voluntary and the autonomy of the respondents was respected. Written informed consent was taken from each participant. The students who were not willing to provide written informed consent and did not complete the questionnaire were excluded from our study.
We used a selfadministered, structured questionnaire which consisted of three parts such as (1) demographic characteristics (age, gender, and ethnicity), (2) attitudes toward statistics, and (3) previous learning experience of mathematics. We utilized SATS36 scale to measure the student's attitudes toward statistics after getting permission from the author.^{[12]} SATS36, a validated instrument, consisted of 36 items which were divided into six subscales such as affect, cognitive competence, value, difficulty, interest, and effort. Affect subscale included six items which measured student's feelings concerning statistics while cognitive competence had also six items which assessed student's attitudes about their intellectual knowledge and skills when applied to statistics. Value subscale consisted of nine items which measured student's attitudes about the usefulness, relevance, and worth of statistics in personal and professional life. Student's attitudes about the difficulty of statistics as a subject were measured in the difficulty subscale which included seven items. Individual's level of interest in statistics was assessed in the interest subscale and the amount of work the student expended to learn statistics was measured in the effort subscale; both subscales included four items. Sevenpoint Likert scale in which one as strongly disagree, four as neither disagree nor agree, and seven as strongly agree, was used. Reverse scoring was done to the negatively worded items in which strongly agree was scored one and strongly disagree was scored seven. The total score of each subscale was computed by taking the sum of the items included and then divided by the number of items within each component. A higher score indicated more positive attitudes toward statistics. The pilot study was conducted with a convenience sample of 30 students to check face validity and internal consistency. Face validity was checked for clarity of wording, the likelihood the target audience would be able to answer the questions, and the layout and style acceptable. The internal consistency of each subscale was calculated using Cronbach's alpha coefficient. The Cronbach's alpha coefficient value of affect subscale was 0.833, cognitive competence was 0.828, value was 0.815, difficulty was 0.643, interest was 0.810, and effort was 0.865. As English is the language of instruction in our college, we did not translate the original English questionnaire to the local language.
After checking and coding the questionnaire, we used PASW Statistics 18 (SPSS Inc., Chicago, Illinois, USA) for data entry and data analysis. For quantitative variables, mean, standard deviation, and range were calculated. Frequency and percentage were described for qualitative variables. We used independent sample ttest to determine the attitudes toward statistics between different genders. We also performed Spearman's rho correlation to find the association between statistics and math cognitive competence, career value using statistics, prior math achievement, and attitudes toward statistics. All the statistical tests were two sided, and the level of significance was set at 0.05.
Confidentiality was maintained and the anonymity of respondents was ensured in this study. In addition, data were kept secured and available only to the statistician. Approval for this study was taken from the Research Ethics Committee, Faculty of Medicine, Melaka Manipal Medical College (MMMC/FOM/Research Ethics Committee – 6/2018).
Results   
A total of 206 students participated in this study. The response rate among medical students was 51.1% (143/280) and dentistry students were 84% (63/75). [Table 1] shows that 69.4% of the respondents were from MBBS program and 30.6% were from BDS program. The mean age was 22.3 years and 61.9% were female students. Among the participants, 86.8% learned statistics before [Table 1].  Table 1: Demographic characteristics among undergraduate medical (n=143) and dentistry students (n=63)
Click here to view 
The mean score in each subscale of attitudes toward statistics, math cognitive competence, career value using statistics, and prior math achievement is shown in [Table 2]. Among subscales of SATS36, the mean score was highest in the effort domain followed by cognitive competence, value, interest, affect, and the lowest mean value was in the difficulty domain [Table 2].  Table 2: Attitudes toward statistics, career value using statistics, math and statistics cognitive competence among undergraduate medical and dentistry students
Click here to view 
[Table 3] shows that there were no significance differences of attitudes toward statistics such as affect (mean difference 0.25 [95% confidence interval (CI) −0.08, 0.59]), cognitive competence (mean difference 0.30 [95% CI −0.01, 0.62]), value (mean difference 0.18 [95% CI −0.08, 0.44]), difficulty (mean difference 0.14 [95% CI −0.09, 0.37]), and effort (mean difference −0.14 [95% CI −0.47, 0.20]) between male and female students. However, male students had a significantly higher mean score in the interest domain (mean difference 0.40 [95% CI: 0.07, 0.74]) [Table 3].  Table 3: Attitudes toward statistics between male and female undergraduate medical and dentistry students
Click here to view 
[Table 4] shows that there were significant positive correlations between career value using statistics, statistics and math cognitive competence, prior math achievement, and attitudes toward statistics in all subscales [Table 4].  Table 4: Correlation between career value using statistics, statistics and math cognitive competence, prior math achievement, and attitudes toward statistics
Click here to view 
Discussion   
This crosssectional study was conducted to determine the undergraduate medical and dentistry student's attitudes toward statistics before the commencement of the biostatistics course. We used SATS36 scale to measure the student's attitudes toward statistics as it has been validated and used in various educational settings and interventions.^{[13],[14],[15],[16],[17],[18]} SATS36 scale measures student's feeling and expression concerning statistics (affect), student's attitudes about their intellectual knowledge and skills when applied to statistics (cognitive competence), student's attitudes about the usefulness, relevance and worth of statistics in personal and professional life (value), student's attitudes toward difficulty in understanding statistics as a subject (difficulty), student's interest in learning and using statistics (interest), and the amount of work the student expended to learn statistics (effort).^{[12]}
We found that the undergraduate medical and dentistry students in our college had positive attitudes toward statistics in most of the domains of SATS36 such as affect, cognitive competence, value, interest, and effort. Among these domains, the effort domain had highest mean value showing the highly positive attitudes of studying statistics, completing the statistics assignments, and attending every statistics class. Previous studies also documented that the students reported spending maximal effort in studying statistics at the beginning of the statistics introductory course.^{[19],[20]} Similar to previous studies,^{[20],[21]} we found that the student's attitudes were more toward negative in the difficulty domain, i.e., the students found statistics is a complicated subject, is difficult to learn, and involves massive computation. Previous research done among 1^{st}year students of the Faculty of Information and Science Technology in Malaysia showed that 71.11% of the students had positive attitudes, but 27.22% had neutral and 1.67% had negative attitudes toward statistics.^{[3]} Contrary to our findings, the same study showed that students had negative attitudes in the components of interest, affective, and effort, but they revealed positive attitudes toward the difficulty component.^{[22]} The study done among undergraduate engineering students in Malaysia displayed positive attitudes toward statistics in affect, cognitive competence, value, interest, and effort constructs, but negative attitudes were exhibited in difficulty construct.^{[21]}
Even though the positive attitudes were displayed among our students, most of our students stated that they were not likely to take statistics courses if the choice had been theirs. It has been reported that statistics is a subject which is difficult and boring,^{[3],[5],[6],[7],[23]} and many students stated that they would not learn unless it is compulsory.^{[7]} Moreover, our students showed neutral attitudes of using statistics in their field of study and future careers which were different from the previous study.^{[22],[23]} Interestingly, we found that the career value using statistics was positively correlated with attitudes toward statistics; the students who intended to use statistics in their future careers were having more positive attitudes about it.
Mixed results had been observed regarding gender differences of attitudes toward statistics. In line with previous studies,^{[3],[5],[21],[24]} our study shows that the attitudes toward statistics were not significant differences between genders. However, we found that male students had significantly higher interest than female students which was not consistent with the previous study.^{[25]} On contrary, Adegboye and Jawid showed that the male students had more positive attitudes toward statistics than their counterparts.^{[2]}
In our study, it was revealed that prior math achievement and math cognitive competence of the students were positively correlated with attitudes toward statistics. Those who reported good mathematics performance previously had shown more positive attitudes toward statistics and vice versa. Previous research studies have shown that the mathematical background of the students and previous mathematics courses taken were related to attitudes toward statistics,^{[5],[17],[26]} though these results were not consistent.^{[27]}
Thus, the positive attitudes toward statistics are fundamental in learning statistics, the instructors need to understand the student's attitudes, how to improve their learning experience, and shape their behavior. The appropriately designed biostatistics course is needed to change student's attitudes toward the desired direction. Not only providing theoretical knowledge, data processing, and analytical skills in statistics classes, it is necessary for the lecturers to create the course enjoyable and more effective by adopting active learning methods such as quizzes, data interpretation exercises, and frequent inclass tests to help the students realize their level of understanding. It is also recommended that the simultaneous research project should be introduced to the statistics course and not in sequence.
There are some limitations in this study. Due to social desirability, the participants may respond in a favorable view to some of the attitude questions. In our study, we were able to invite only three groups of senior year students in the academic year 3 of the MBBS program due to the final examination. This study was conducted in one private institution; therefore, the findings cannot be applicable to other institutions nor other settings. This study can be replicated by recruiting a large number of students from different medical institutions ensuring high generalizability findings. Emphasis should also be given to clinicians as there is a growing need to interpret effectively the epidemiological data and results in medical literature. Therefore, future research should also investigate the anxiety and attitudes toward statistics among clinicians in both the medical and dentistry fields.
Conclusions   
The undergraduate medical and dentistry students in our college had positive attitudes toward statistics, but the students found that statistics is a complicated subject and is difficult to learn. Furthermore, most of our students stated that they were not likely to take statistics courses if the choice had been theirs. Hence, it is important for the instructors to understand the student's attitudes and create effective learning strategies which not only provide knowledge and skills but also change student's attitudes toward the desired direction.
Ethical clearance
This study was approved by Research Ethics Committee, Faculty of Medicine, Melaka Manipal Medical College (MMMC), Malaysia (MMMC/FOM/Research Ethics Committee – 6/2018).
Acknowledgment
The authors would like to acknowledge to all students who participated in this study. We also would like to thank the management of MMMC and the Research Ethics Committee, FOM, MMMC to grant the approval of this study.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  Daniel WW, Cross CL. Biostatistics: A Foundation for Analysis in the Health Sciences. 10 ^{th} ed. USA: John Wiley and Sons; 2013. 
2.  Adegboye OA, Jawid A. Multivariate multilevel models for attitudes toward statistics: Multidisciplinary settings in Afghanistan. J Appl Stat 2016;43:44261. 
3.  Judi HM, Sahari N, Mohamed H, Wook TM. Students profile based on attitude towards statistics. Procedia Soc Behav Sci 2011;18:26672. 
4.  Altman DG. Practical Statistics for Medical Research. London: Chapman and Hall; 1991. 
5.  Zhang Y, Shang L, Wang R, Zhao Q, Li C, Xu Y, et al. Attitudes toward statistics in medical postgraduates: Measuring, evaluating and monitoring. BMC Med Educ 2012;12:117. 
6.  Kiekkas P, Panagiotarou A, Malja A, Tahirai D, Zykai R, Bakalis N, et al. Nursing students' attitudes toward statistics: Effect of a biostatistics course and association with examination performance. Nurse Educ Today 2015;35:12838. 
7.  Gordon S. Understanding students' experiences of statistics in a service course. Stat Educ Res J 2004;3:4059. 
8.  Schau C. Students' Attitudes: The “Other” Important Outcome in Statistics Education. Proceedings of the Joint Statistical Meetings; August, 2003. p. 367381. 
9.  Liau AK, Kiat JE, Nie Y. Investigating the pedagogical approaches related to changes in attitudes toward statistics in a quantitative methods course for psychology undergraduate students. Asia Pac Educ Res 2015;24:31927. 
10.  Kyaw Soe HH, Than NN, Lwin H, Nu Htay MN, Phyu KL, Abas AL. Knowledge, attitudes, and barriers toward research: The perspectives of undergraduate medical and dental students. J Educ Health Promot 2018;7:23. 
11.  Lwanga SK, Lemeshow S. Sample Size Determination in Health Studies: A Practical Manual. Geneva: World Health Organization; 1991. 
12.  Schau C, Stevens J, Dauphinee TL, Vecchio AD. The development and validation of the survey of antitudes toward statistics. Educ Psychol Meas 1995;55:86875. 
13.  Cashin SE, Elmore PB. The survey of attitudes toward statistics scale: A construct validity study. Educ Psychol Meas 2005;65:50924. 
14.  Chiesi F, Primi C. Assessing statistics attitudes among college students: Psychometric properties of the Italian version of the Survey of Attitudes toward Statistics (SATS). Learn Individ Dif 2009;19:30913. 
15.  Carlson KA, Winquist JR. Evaluating an active learning approach to teaching introductory statistics: A classroom workbook approach. J Stat Educ 2011;19:123. 
16.  Carnell LJ. The effect of a studentdesigned data collection project on attitudes toward statistics. J Stat Educ 2008;16:1. [doi: 10.1080/10691898.2008.11889551]. 
17.  Dempster M, McCorry NK. The role of previous experience and attitudes toward statistics in statistics assessment outcomes among undergraduate psychology students. J Stat Educ 2009;17:2. [doi: 10.1080/10691898.2009.11889515]. 
18.  Posner MA. The impact of a proficiencybased assessment and reassessment of learning outcomes system on student achievement and attitudes. Stat Educ Res J 2011;10:314. 
19.  Millar AM, Schau C. Assessing Students' Attitudes: The Good, the Bad, and the Ugly. Proceedings of the Joint Statistical Meetings; 2010. p. 113343. 
20.  Bond ME, Perkins SN, Ramirez C. Students' perceptions of statistics: An exploration of attitudes, conceptualizations, and content knowledge of statistics. Stat Educ Res J 2012;11:625. 
21.  Ghulami HR, Ab Hamid MR, Zakaria R. Comparative analysis of positive and negative attitudes toward statistics. AIP Conf Proc 2015;1643:2128. 
22.  Ashaari NS, Judi HM, Mohamed H, Wook MT. Student's attitude towards statistics course. Procedia Soc Behav Sci 2011;18:28794. 
23.  Wu Y, Zhang L, Liu L, Zhang Y, Liu X, Yi D. Attitudes of medical graduate and undergraduate students toward the learning and application of medical statistics. J Biol Educ 2015;49:22031. 
24.  Cherian VI, Glencross MJ. Sex, socioeconomic status, and attitude toward applied statistics among postgraduate education students. Psychol Rep 1997;80 3 Suppl: 13856. 
25.  Fullerton JA, Kendrick A. The math problem: Advertising students' attitudes toward statistics. Journalism Mass Commun Educ 2013;68:1349. 
26.  Carmona J, Martínez RJ, Sánchez M. Mathematical background and attitudes toward statistics in a sample of Spanish college students. Psychol Rep 2005;97:5362. 
27.  Scott JS. Modeling Aspects of Students' Attitudes and Performance in an Undergraduate Introductory Statistics Course [Dissertation]. Athens GA: University of Georgia; 2001. 
[Table 1], [Table 2], [Table 3], [Table 4]
