Development and internal validation of an instrument to measure the motivation of residents for family medicine

Abstract

Background

For several decades, medical school graduates’ motivation to specialise in family medicine is decreasing. Therefore, residents in family medicine must be motivated for the profession and finish their residency.

Objectives

Goal of the current study is the development and internal validation of an instrument to measure the residents’ motivation for family medicine, which is based on the self-determination theory: STRength mOtivatioN General practitioner (STRONG).

Methods

We used an existing instrument, the ‘Strength of Motivation for Medical School,’ adapted the 15 items and added a 16th item to make it suitable for residency in family medicine. After a review by experts, the questionnaire was sent to 943 residents of family medicine in Bavaria, Germany, in December 2020. An exploratory factor analysis for the STRONG item scores was carried out. The items were analysed for grouping into subscales by using principal component analysis. Cronbach’s alpha for internal consistency was determined for calculating the reliability of the subscales.

Results

After analysis, the questionnaire appeared to consist of two subscales: ‘Willingness to sacrifice’ (eight items, Cronbach’s alpha is 0.82) and ‘Persuasion’ (five items, Cronbach’s alpha is 0.61). The factor analysis with Promax rotation resulted in two factors explaining 39.6% of the variance. The Cronbach’s alpha of the full scale is 0.73.

Conclusion

Based on the internal validation, the STRONG Instrument appears to have good reliability and internal validity, assuming a two-factor structure. This may therefore be a helpful instrument for measuring the strength of the motivation of (future) family medicine residents.

Keywords:

• Motivation
• primary care
• family medicine
• residents
• validation

KEY MESSAGES

• STRONG may be a valuable instrument to measure the motivation of (future) residents in family medicine.

• One possible application of the instrument is that by measuring motivation at several points during residency, the factors that influence motivation can be identified.

• It is recommended to validate this tool also for other countries and contexts.

Introduction

For several decades, the motivation of medical school graduates to specialise in family medicine has been decreasing and in many countries smaller than the demand for general practitioners [1–4]. Moreover, residents in family medicine must be motivated for the profession, which increases the chance that they will finish residency, work in family medicine with pleasure for a long time and continue to work on developing their professional skills throughout their careers [5–7].

A commonly used definition of motivation is ‘Powering people to achieve high levels of performance and overcoming barriers to change’ [8]. Motivation is a feeling that moves people to do something and is related to purposefulness and success [5,9].

The self-determination theory (SDT) argues that motivation is influenced by three psychological needs: autonomy, relatedness and competence. SDT differentiates among more or less autonomous motivation, intrinsic, extrinsic and amotivation [10,11]. Intrinsic motivation reflects engagement in an activity due to inherent interest. It is an autonomous type of motivation. Extrinsic motivation refers to engagement in an activity to obtain outcomes, like to gain rewards or social approval, avoid punishment or comply with external norms. They are considered controlling types of motivation. Motivational regulations have different cognitive, affective and behavioural consequences [10,11].

There is a direct relevance from educational settings to intrinsic and extrinsic motivation [12]. As Kusurkar et al. describe in their review [13], motivation in medical education can be both a dependent and an independent variable. In the function of dependent variable, motivation can be influenced by changes in the curriculum or the learning environment in clinical practice. Motivation as an independent variable can influence students’ learning behaviour and study success [13].

Motivation is a major determinant of the quality of learning, the learning strategies students use, their persistence and performance [8,10,14]. The hypothesis is that motivated students try harder during the lesson, pay more attention and persist more to learn the skills than the less intrinsically motivated students. It confirms that autonomous forms of motivation positively influence performance [10,14].

Several validated questionnaires measure the strength of motivation of medical students [15–17]. This makes it possible, for example, to measure changes in motivation over time or to measure correlations between motivation and other factors, such as empathy or personality traits [11,18,19].

Many studies have investigated factors and motivations impacting the decision to choose specific medical specialties. Motivations may include lifestyle choices, the possibility of private practice, interest in particular diseases, motivation for research or teaching or to gain a higher income. Students’ speciality preferences are influenced mainly by interaction with patients and desire to work in a challenging speciality while less common reasons are high demand for specific health services, the income potential, the workload, prestige or the duration of the residency [16,20,21].

Studies on the motivation for a specialisation generally aim to inventory the students’ or residents’ reasons for a particular choice. These are usually self-designed questionnaires, sometimes supplemented with interviews [2,3,20]. Due to its primarily qualitative nature, these questionnaires cannot be used to measure the strength of motivation.

Goal of the current study therefore is the development and internal validation of an instrument to measure the motivation of residents for family medicine: STRength mOtivatioN General practitioner (STRONG).

Methods

Initial instrument

For the development of the STRONG questionnaire, we have chosen to use an existing instrument as a starting point, namely the SMMS: Strength of Motivation for Medical School [15]. This questionnaire is developed to measure the strength of motivation of medical students for medical school. This instrument based on the Self-Determination Theory by Ryan and Deci [9,12], has shown good psychometric properties. The SMMS-questionnaire consists of 15 items scored on a five-point Likert scale, from ‘strongly disagree’ tot ‘strongly agree’. The higher the score, the stronger the motivation. This instrument consists of three sub-scales: ‘willingness to sacrifice’, ‘readiness to start’ and ‘persistence’. The internal consistency of the three subscales, with each five items, measured by Cronbach’s alpha, ranges from 0.55 to 0.67 [15].

Development STRONG

To stay as close as possible to the original questionnaire, we used the Dutch version of the SMMS. We adapted the 15 items to make them suitable for family medicine residency and added a 16th item. This new questionnaire was translated back and forth from Dutch into German and English by native speakers. The items were then critically reviewed in two rounds by general practitioners involved in residency training (n = 4) and family medicine residents (n = 3). Based on the comments from these panels, the items were slightly edited. The items of the SMMS and the STRONG that we used in this validation study are listed in Table 1.

Table 1. Items SMMS and STRONG.

	SMMS: Strength of motivation for medical school (with subscale)	STRONG: STRength mOtiviatioN General Practitioner
1	I would always regret my decision if I hadn’t availed myself of the opportunity to study medicine (2).	I would always regret it if I did not take the opportunity to do a residency programme in family medicine (2).
2^a	I would quit studying medicine if I were 95% certain that I could never become the specialist of my choice (3).	I would discontinue residency if I were 95% sure that I couldn’t work in family medicine (–).
3	I would still choose medicine even if that means studying in a foreign country in a language I have not yet mastered (2).	I would also like to start or continue residency in family medicine if I could only do this abroad in a language I do not speak at the moment (1).
4^a	As soon as I discovered that it would take me 10 years to qualify as a doctor, I stopped studying. (3)	As soon as I discovered that it would take 10 years to get a specialist degree in family medicine, I would discontinue residency (2).
5	Even if I could not maintain my social life, I would continue medical training (1).	Even if I could hardly maintain my social life, I would continue residency in family medicine (1).
6	I wouldn’t consider any other profession than becoming a doctor (2).	I wouldn’t consider any other profession than becoming a general practitioner (2).
7	I would still choose medicine even if that meant I could never go on holidays with my friends anymore (1).	I would also choose to become a general practitioner even if that meant I would never be able to go on holiday with my friends or family anymore (1).
8^a	I would stop studying medicine if I started scoring low marks and failing tests often (3).	I would discontinue residency in family medicine if it turned out that I kept getting negative feedback about the way I work (1).
9^a	If studying took me more than an average of 60 h a week, I would seriously consider quitting (1).	If the residency in family medicine would cost me more than an average of 60 h of work per week all year round, I would seriously consider dropping out of residency (1).
10	I intend to become a doctor even though that would mean taking CME courses two evenings a week throughout my professional career (1).	I intend to become a general practitioner, even though that would mean taking CME courses two evenings a week throughout my professional career (–).
11^a	It wouldn’t bother me too much if I could no longer study medicine (2).	It wouldn’t bother me too much if I could not become a general practitioner (any more) (2).
12	I would like to become a doctor, even if that means giving precedence to my work over my family (1).	I want to become a general practitioner if that would mean giving precedence to my work over my family (1).
13^a	I would quit studying as soon as it became apparent that there were no jobs or resident positions after graduation (3).	I would discontinue residency as soon as it became apparent that there were no jobs in family medicine (–).
14^a	I would not have chosen medicine if it had caused me to accumulate substantial financial debts (3).	I would not have chosen family medicine if it had caused me to accumulate substantial financial debts (1).
15	I would be prepared to retake my final high school exams to get higher marks if this would be necessary to study medicine (2).	I would be prepared to retake my state exams to achieve a higher average mark if this would be necessary to become a specialist in family medicine (1).
16^a		As soon as I discovered that I had to go to a rural region to become a general practitioner, I would discontinue residency (2).

Notes: Subscales SMMS: 1 = willingness to sacrifice, 2 = readiness to start, 3 = persistence. Subscales STRONG: 1 = willingness to sacrifice, 2 = persuasion.

^aTo be reverse scored.

Internal validation

The STRONG questionnaire was sent electronically to 943 family medicine residents in Bavaria, Germany, in December 2020. A reminder was sent two weeks after the original invitation.

For the statistical analyses, the software programme SPSS version 27 was used. An exploratory factor analysis for the STRONG item scores was carried out. The items were analysed for grouping into subscales by using principal component analysis. Cronbach’s alpha for internal consistency was determined for calculating the reliability of the subscales.

Ethical approval

Approval for the study was obtained from the Ethics Committee of the Faculty of Medicine of the Technical University of Munich (approval no. 627/19S). The survey was anonymous and voluntary, and all participants gave written consent in accordance with the Declaration of Helsinki. All residents received information on survey’s nature, purpose and procedure and their right to withhold or revoke their consent at any time.

Results

Of the 943 family medicine residents, 275 completed the questionnaire making a response rate of 29.1%. The gender distribution was 83.3% females and 16.7% males. In the entire research group, 78.3% are women meaning that women are slightly over-represented among the respondents. The average age of the responders was 36.7 years, corresponding to the average age in the whole study group (36.8).

The value of the KMO (Kaiser, Meyer und Olkin) test was .79 and the Bartlett Test resulted in a value of < 0.001, indicating that the data was suitable for a factor analysis [22]. The factor analysis with Promax rotation resulted in two factors (see screeplot – Figure 1) explaining 39.6% of the variance [22].

Figure 1. Screeplot of the factor analysis. The x-axis shows the questionnaire items. The y-axis shows the eigenvalues per item.

The questionnaire subscales could be labelled as: ‘Willingness to sacrifice’ (Subscale1) and ‘Persuasion’ (Subscale 2).

Subscale 1: Willingness to sacrifice

This subscale measures to what extent residents are prepared to make sacrifices to become general practitioners. Eight items based on their factor loadings (> 0.40) belonged to this subscale – see Table 2 [23]. This subscale explained 25.2% of the variance. Although item 2 had a factor loading of only 0.32, we checked the effect of adding this factor on the internal consistency of the subscale. The Cronbach’s alpha of this subscale was 0.82 and decreased to 0.81 when item 2 was added, so we decided to exclude the item from the subscale.

Table 2. Factors loadings of the STRONG items.

No.	STRONG item	Factor 1	Factor 2
3	I would also like to start or continue residency in family medicine if I could only do this abroad in a language I do not speak at the moment.	0.52	0.01
5	Even if I could hardly maintain my social life, I would still continue residency in family medicine.	0.70	0.08
7	I would also choose to become a general practitioner even if that meant I would never be able to go on holiday with my friends or family anymore.	0.82	0.07
8	I would discontinue residency family medicine if it turned out that I kept getting negative feedback about the way I work.	0.40	0.09
9	If the residency in family medicine would cost me more than an average of 60 h of work per week all year round, I would seriously consider dropping out of residency.	0.57	0.10
12	I want to become a general practitioner if that would mean giving precedence to my work over my family.	0.68	0.09
14	I would not have chosen family medicine if it had caused me to accumulate substantial financial debts.	0.54	0.05
15	I would be prepared to retake my state exams to achieve a higher average mark if this would be necessary to become a specialist in family medicine.	0.60	0.06
1	I would always regret it if I did not take the opportunity to do a residency programme in family medicine.	0.06	0.69
4	As soon as I discovered that it would take 10 years to get a specialist degree in family medicine, I discontinued residency.	0.15	0.28
6	I wouldn’t consider any other profession than becoming a general practitioner	0.31	0.45
11	It wouldn’t bother me too much if I could not become a general practitioner (any more).	0.17	0.76
16	As soon as I discovered that I had to go to a rural region to become a general practitioner, I discontinued residency.	0.26	0.46
	Deleted items
2	I would discontinue residency if I were 95% sure that I couldn’t work in family medicine.	0.32	0.21
10	I intend to become a general practitioner, even though that would mean taking CME courses two evenings a week throughout my professional career.	0.13	0.23
13	I would discontinue residency as soon as it became apparent that there were no jobs in family medicine.	0.23	0.10

Bold indicates to which factor the respective item was assigned.

Subscale 2: Persuasion

The subscale ‘persuasion’ measures how convinced the residents are of their choice for family medicine. This subscale explained 14.4% of the variance. Four items belonged to this subscale based on the factor loadings of > 0.40 (see Table 2). Although item 4 had a factor loading of only 0.28, we decided to include it into the subscale because it improved the internal consistency of the subscale. Cronbach’s alpha of the subscale was 0.61, which is acceptable.

There were no cross loadings of items in one subscale onto the other subscale with a value of > 0.40. The Cronbach’s alpha for the full scale, with 13 items was 0.73, which is good.

Discussion

Main findings

As far as we know, the STRONG-questionnaire, based on the Self-Determination Theory by Ryan and Deci [9,12], is the only existing instrument for measuring the strength of motivation of residents for family medicine. Based on the internal validation described in this paper, this instrument appears to have good reliability and internal validity, assuming a two-factor structure. The reliability for the ‘willingness to sacrifice’ subscale and ‘persuasion’ subscale are very good (0.82) and acceptable (0.61), respectively. The reliability of the total instrument with 13 items is 0.73.

Comparison SMMS – STRONG

The SMMS, on which the STRONG is based, has three subscales [15]. The questions in the ‘Willingness to sacrifice’ subscale largely overlap with the modified questions on scale 1 in our instrument. Therefore, we have given it the same name. The other questions fit our second subscale. We cannot confidently say why we found two rather than three subscales. The most likely explanation is that it has to do with the different study group (students versus residents) and different content (medicine in general versus family medicine).

Implications of the internal validation

Our study findings mean that the STRONG instrument may be used to measure the strength of the motivation of family medicine residents at different moments during residency: by regularly evaluating the strength of motivation during residency, factors which influence motivation positively or negatively can be identified [13]. This can be done, for example, by measuring the motivation of cohorts of residents over a longer period and seeing how motivation develops. If an apparent increase or decrease can be seen at certain moments, it will be worthwhile to see what possible factors may have contributed. This is especially relevant because in addition to the primary motivation, multiple aspects such as work environment and professional development opportunities, can impact motivation for family medicine [1].

Furthermore, insights into the strength of motivation for the profession are also important later on, as motivation affects the extent to which general practitioners continue to take courses throughout their careers, which increases their employability [24]. To gain insight into the usefulness of this instrument, for example, whether low motivation strengths lead to a higher risk of dropping out, the instrument will have to be used over a longer period. Other possible applications for this tool include calculating the correlation with other factors, such as professional identity formation regarding family medicine or personality traits [25,26]. We do not want to promote using the instrument as a selection tool for family medicine residency or in other situations that may encourage socially desirable responses. The questionnaire has not been validated for selection purposes.

Moreover, this instrument may also be used during medical school, for example, for students considering taking part in a special track that prepares them for family medicine [23]. The validation of this instrument for this group could be done in a follow-up study.

As described earlier, motivation can be more or less intrinsic or extrinsic, with intrinsic and autonomous forms of motivation in particular positively affecting performance and perseverance [8,10,14]. The SMMS, on which our instrument is based, appears to correlate especially positively with intrinsic motivation [15]. A relevant follow-up study is to examine if the same is true for the STRONG.

Strengths and limitations

This study has several limitations. The most significant limitation is the moderate response rate. Because of that, we cannot exclude a selection bias, meaning that more motivated residents may be more inclined to respond. The number of female respondents is slightly higher than the percentage of women among the Bavarian family medicine residents but as the difference is small, it does not affect generalisability. Besides, the sample of this internal validation study only consisted of German residents. Although there are no indications that this target group differs greatly from residents in family medicine in other countries, it is recommended that the validation study be repeated in several countries and different contexts (for example, with residents in other disciplines) for external validation, supplemented by applying the tool in different contexts to test for feasibility.

Conclusion

Based on the internal validation, the STRONG Instrument appears to have good reliability and internal validity, assuming a two-factor structure. This may therefore be a useful instrument for measuring the strength of the motivation of (future) family medicine residents. Insight into the motivation of family medicine residents may increase the likelihood that they will complete their training and stay in family medicine for a long time.

Ethical approval

Approval for the study was obtained from the Ethics Committee of the Faculty of Medicine of the Technical University of Munich (approval no. 627/19 S). The survey was anonymous and voluntary. All students received information on the nature, purpose and procedure of the survey and their right to withhold or revoke their consent at any time.

Consent form

All participants have written Informed Consent prior to the study. All methods were carried out in accordance with Declaration of Helsinki.

Acknowledgements

We would like to thank Dagmar Scheider from KOSTA for sending the questionnaire to the residents.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Data availability statement

The datasets generated and/or analysed during the current study are not publicly available due to data protection guidelines of the institution but are available from the corresponding author upon reasonable request.