|
 
 |
| ORIGINAL ARTICLE |
|
| Year : 2019 | Volume
: 6
| Issue : 1 | Page : 7-11 |
|
Assessment of reliability and adaptation of fisher's 52-item self-directed learning readiness scale among medical students in Southern India
Sujiv Akkilagunta1, Sitanshu Sekhar Kar2, KC Premarajan2, Subitha Lakshminarayanan2, Archana Ramalingam3, Thomas V Chacko4, Shital Bhandary5
1 Department of Community Medicine, All India Institute of Medical Sciences, Nagpur, Maharashtra, India
2 Department of Preventive and Social Medicine, Jawaharlal Institute of Postgraduate Medical Education and Research, Puducherry, India
3 Department of Epidemiology, Institute of Liver and Biliary Sciences, New Delhi, India
4 Department of Community Medicine, Believers Church Medical College and Hospital, Kerala, India
5 Department of Public Health and Medical Education, Patan Academy of Health Sciences, Lalitpur, Nepal
| Date of Submission | 02-Aug-2018 |
| Date of Acceptance | 27-Mar-2019 |
| Date of Web Publication | 10-Jul-2019 |
Correspondence Address:
Dr. Sitanshu Sekhar Kar
Department of Preventive and Social Medicine, JIPMER, Puducherry
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/IJAMR.IJAMR_39_18
Background: The concept of self-directedness is defined by the active participation of students in all phases of learning, from framing the learning objectives to the assessment process. Various scales have been developed to measure self-directed learning (SDL) among students. Fishers' SDL readiness scale (SDLRS) was validated among nursing students. The present study assessed the reliability of Fisher's SDLRS among medical students in India. Factor analysis was carried out to reduce the total number of items in the scale. Materials and Methods: The SDLRS used in the study consisted of 52 items classified under three domains. The scale was administered to a sample of 367 medical students between the first and seventh semesters. Cronbach's alpha was used to measure the reliability of the scale. For item-reduction and further abridgment, factor analysis was done using principal component analysis. Results: Cronbach's alpha was found to be 0.93 for Fisher's 52-item scale. Satisfactory reliability was observed across each domain (≥0.7). Factor analysis enabled the reduction of scale to 29 items with three domains. Conclusion: Fisher's 52-item SDLRS is a reliable scale for use among medical students in India. We propose the use of a reliable and shortened 29-item scale.
Keywords: Medical students, reliability, self-directed learning, self-directed learning readiness scale
How to cite this article:
Akkilagunta S, Kar SS, Premarajan K C, Lakshminarayanan S, Ramalingam A, Chacko TV, Bhandary S. Assessment of reliability and adaptation of fisher's 52-item self-directed learning readiness scale among medical students in Southern India. Int J Adv Med Health Res 2019;6:7-11 |
How to cite this URL:
Akkilagunta S, Kar SS, Premarajan K C, Lakshminarayanan S, Ramalingam A, Chacko TV, Bhandary S. Assessment of reliability and adaptation of fisher's 52-item self-directed learning readiness scale among medical students in Southern India. Int J Adv Med Health Res [serial online] 2019 [cited 2023 Apr 2];6:7-11. Available from: https://www.ijamhrjournal.org/text.asp?2019/6/1/7/262491 |
| Introduction | |  |
Adult learning principles have gained impetus in the past three to four decades. Knowles has described that the concept of adult learning as the assumption that a person grows from being totally dependent toward what is termed as self-directedness.[1] It is a behavioral change where the student takes the initiative in framing his/her own learning objectives, forms action plans, and carries them out.
Student-centered learning is an approach based on these principles. In this approach, the student takes the initiative in planning and executing the learning activities. The role of a teacher here is more of facilitation rather than an authoritative one.[2],[3] It indicates the preparedness of the student to take active role in his own learning. Hence, before initiating new approaches based on adult learning, it becomes crucial to measure a student's self-directed learning (SDL).
To measure a learner's SDL, Gugleilmino devised a 58-item scale. The instrument consisting of eight domains was used most often to measure SDL.[4] Fisher et al. developed another scale to measure SDL among nursing students with 93 items in the pilot phase. This was reduced to 52 items by Delphi method and later to 40 items with three major domains, namely, “self-management,” “desire for learning,” and “self-control.”[5] The validity of the scale was assessed among medical students in Chile using exploratory factor analysis.[6] Since it has been validated in several studies, validation of the scale was not attempted in the present study.
A lengthy questionnaire with 40 items not only reduces the response rate but also increases the response duration.[7] Reduction in the number of items could reduce the length of the scale without altering the conceptual framework of the scale. Although Fisher's scale had been used among medical undergraduates in India,[8] the reliability of the scale was not assessed. Hence, the following study was conducted to abridge the scale and assess its reliability.
In this study, we assessed the reliability of the 52-item SDL readiness scale (SDLRS) in the Indian context. Factor analysis was conducted using principal component analysis (PCA) to facilitate item reduction in the SDLRS.
| Materials and Methods | | |
Study setting
Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER) is an institution of national importance under the Ministry of Health and Family Welfare, Government of India, located in Puducherry. Medical students are admitted to JIPMER through a nation-level entrance test. During the period of study, this number increased from 100 to 150. The MBBS curriculum consists of nine semesters each of 6 months' duration followed by 1 year of internship. Community medicine is taught as a part of the curriculum that begins in the first semester and ends in the seventh semester.
Ethical considerations
The study was approved by the Institute Ethics Committee. Written informed consent was obtained from each student before administering the questionnaire.
Data collection
Data were collected as a part of a research project on implementing a student-centered approach in learning, conducted among students in the Community Medicine class. In the project, the students were divided into groups and were given the responsibility to teach their own classmates on selected topics from the subject. The students chose the learning objectives, teaching methods, and assessment process. The faculty played the role of facilitators during the course of the project.
To assess the SDL, a 52-item self-administered SDLRS questionnaire was administered after obtaining informed consent from all the participants. The students were explained the purpose of the study and instructed regarding the questionnaire. Anonymity was permitted among participants. Data were collected on sociodemographic variables such as age, gender and area of residence, place of stay (hostel or day scholar), board of education, and medium of instruction during their secondary and higher secondary education as well as information on whether there are any doctors in the family. Responses were collected from 367 students who were between the first and seventh semesters at the time of the study.
Self-directed learning readiness scale questionnaire
Fisher's SDLRS is a self-administered questionnaire consisting of 40 items under three domains – self-control, self-management, and desire for learning. The items are scored on a Likert scale from 1 to 5. The unabridged version with 52 items was used for the purpose of the study to facilitate abridgment of the scale. The total score for the unabridged version ranges from 52 to 260.
Statistical analysis
The data were entered into MS Excel 2010 and analyzed using IBM SPSS version 20 (IBM SPSS Statistics for Windows, IBM Corp., Armonk, NY, USA).[9] To assess the internal consistency-reliability, Cronbach's alpha (α) was measured. Cronbach's alpha value of ≥0.7 was considered satisfactory.[10] To test sampling adequacy for factor analysis, Kaiser-Meyer-Olkin (KMO) and Bartlett's test of sphericity were done. For KMO test, a value of ≥0.6 was considered significant. In Bartlett's test of sphericity, a 5% of significance level was applied. Factorial structure was determined by PCA with Promax rotation. Items with factor loading of ≥0.5 were considered for inclusion in the final adapted scale.
| Results | | |
Data quality
There were 367 students from different batches. Nonresponse in the study was 13.4% (n = 57) and varied from as low as 3.4% in the first semester to as high as 26% in one of the seventh-semester batches.
Item nonresponse was found in the study with 26.4% of the questionnaires (n = 97) having one or more items missing. Of all the 52 items, there were seven items which had an item nonresponse of >2%.
Sample characteristics
Among the 367 students, the majority belonged to an urban area (86.4%) and did their higher secondary education in English medium (95.2%) [Table 1]. More than two-fifths (44.1%) of the students received schooling from the respective state board of education during preuniversity education. A significant proportion of them (40.1%) had doctors among their relatives. | Table 1: Sociodemographic characteristics of the study participants (n=367)
Click here to view |
Self-directed learning readiness scale score
The possibility of total SDLRS score ranged from 52 to 260. The median total score (interquartile range) in the study group was 192 (175–209).
Internal reliability
The scale was analyzed domain-wise and overall for its internal consistency-reliability. Overall Cronbach's alpha for the 52-item scale was found to be 0.93 which indicates high reliability. Domain-wise Cronbach's alpha is presented in [Table 2]. Sensitivity analysis performed after excluding those items with >2% missing responses did not change the value of Cronbach's alpha. Reliability was assessed among different subgroups. The reliability was good (≥0.9) among different subgroups based on the gender, area of residence, and area of current stay. | Table 2: Reliability assessment of Fisher's 52-item self-directed learning readiness scale and the abridged 29-item scale
Click here to view |
Factorial structure
KMO measure of sampling adequacy was found to be 0.88 which is considered meritorious.[11] Bartlett's test of sphericity done on the sample revealed a P < 0.001 (χ2 = 6134.14, df = 1326). Both the statistical tests indicated the adequacy of the sample for factor analysis.
PCA with Promax rotation was done to analyze the factorial structure of the scale. Since the original scale had three domains, factor extraction was limited to three factors. These three factors had an Eigenvalue of >2 and contributed to 35.8% of the total variance. Items with a factor loading of ≥0.5 were included resulting in abridged scale with 29 items [Table 3]. The factorial structure of the scale along with the factor loadings is presented in [Table 4].
The final abridged scale was analyzed by domains for its internal consistency-reliability which showed satisfactory reliability in all domains (Cronbach's alpha ≥0.7) [Table 2].
| Discussion | | |
The 52-item scale used among Indian medical students showed good reliability with a Cronbach's alpha of 0.93. Reliability was above 0.7 across each domain. The subgroup analysis confirmed the reliability of 52-item scale in subgroups of gender, residential area, and area of current stay.
Cronbach's alpha in the present study (0.93) was comparable to that of the study by Fisher et al. (0.92). A similar value was observed (0.90) for the 40-item scale among Chilean medical students by Fasce et al.
Using PCA, the scale was reduced to 29 items with three domains. In factor analysis, the three factors contributed to 35.8% of the variance similar to that of the study conducted by Fisher et al. (36.4%). Abraham et al. conducted a study among 1st-year Indian medical undergraduates, in which they used the 40-item scale by Abraham et al.[8] The 52-item scale was used to facilitate factor analysis. We did not attempt reverse scoring of items as done by Abraham et al. to avoid confusion owing to the use of a lengthier scale (52 vs. 40 items).
Factor analysis was attempted with a predetermined objective to reveal three domains to correspond with domains in the original scale. Factor loading of 0.3 was used as a cutoff in Fisher's scale for nursing students. In comparison, the present study used a stronger factor loading of 0.5 to define items in each domain.
The abridged scale showed good overall reliability (Cronbach's alpha = 0.87). All three domains had satisfactory reliability (≥0.7). Apart from reducing respondent fatigue, the shortened version could reduce item nonresponse that was evident in this study.
The study could have been subject to response bias. Due to the use of a lengthy scale, there is a possibility for acquiescence bias.[12] However, since the respondent was free to remain anonymous, other biases such as social desirability bias, faking good, and faking bad are expected to be minimum.
The medical students in JIPMER are selected by a rigorous nation-wide test and may not be representative of all the medical students in India which could limit the external validity of the study. However, satisfactory reliability demonstrated across the different subgroups in the study could address the issue of generalizability of the study results to some extent.
| Conclusion | | |
Fisher's SDLRS is a reliable scale for measuring SDL. A reliable abridged 29-item version of the scale was developed using factor analysis. We propose the use of this abridged 29-item scale for measuring SDL among medical students to improve the feasibility and reduce nonresponse.
Acknowledgment
We are grateful to Dr. Gautam Roy, Professor, Dr. Iswarya S, Junior resident, Department of Preventive and Social Medicine, JIPMER, for their support throughout the study. We are also grateful to the students for participating in this study.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Knowles M. Appendix D. Toward a model of lifelong education. The Adult Learner: A Neglected Species. Houston, TX: Gulf Publishing Company; 1973. p. 161.  |
| 2. | Ang RP, Gonzalez M, Liwag D, Santos B, Vistro-Yu C. Elements of Students Centered Learning. Philippines: Office of Research and Publications, Loyola School, Ateneo de Manila University; 2001. |
| 3. | Canon R. Guide to Support the Implementation of the Learning and Teaching Plan Year. Adelaide: ACUE, the University of Adelaide; 2000. |
| 4. | |
| 5. | Fisher M, King J, Tague G. Development of a self-directed learning readiness scale for nursing education. Nurse Educ Today 2001;21:516-25. |
| 6. | Fasce HE, Pérez VC, Ortiz ML, Parra PP, Matus BO. Factorial structure and reliability of Fisher, King & Tague's self-directed learning readiness scale in Chilean medical students. Rev Med Chil 2011;139:1428-34. |
| 7. | Jepson C, Asch DA, Hershey JC, Ubel PA. In a mailed physician survey, questionnaire length had a threshold effect on response rate. J Clin Epidemiol 2005;58:103-5. |
| 8. | Abraham RR, Fisher M, Kamath A, Izzati TA, Nabila S, Atikah NN, et al. Exploring first-year undergraduate medical students' self-directed learning readiness to physiology. Adv Physiol Educ 2011;35:393-5. |
| 9. | IBM Corp. IBM SPSS Statistics for Windows. Ver. 20.0. Armonk, NY:IBM Corp; 2011. |
| 10. | Bland JM, Altman DG. Cronbach's alpha. BMJ 1997;314:572. |
| 11. | Kim JO, Mueller CW. Factor Analysis: Statistical Methods and Practical Issues. Beverly Hills, CA: SAGE; 1978. p. 92. |
| 12. | Couch A, Keniston K. Yeasayers and naysayers: Agreeing response set as a personality variable. J Abnorm Soc Psychol 1960;60:151-74. |
[Table 1], [Table 2], [Table 3], [Table 4]
|
Search Pubmed for