Validity and reliability of the Finnish Motivation for Physical Activity (RM4-FM) questionnaire CURRENT STATUS: UNDER REVIEW

Background There is a lack of properly validated instruments measuring motivation for physical activity in the Finnish language. The study aimed to translate the Motivation for Physical Activity (RM4-FM) instrument into Finnish and examine its psychometric properties in a sample of healthy, older Finnish adults. Methods The RM4-FM was translated and linguistically validated adhering to published guidelines. The sample consisted of 102 65–75-year old participants, who completed the RM4-FM, the Physical Activity Acceptance Questionnaire (PAAQ) and sociodemographic characteristics questionnaires electronically. The RM4-FM was re-administered one month after the initial assessment. Results RM4-FM translated well into Finnish. The motivation dimension scores were skewed towards high internal and low external motivation. A floor effect was confirmed in the External regulation dimension and a ceiling effect was confirmed in the Identified regulation dimension. Confirmatory factor analysis indicated inadequacy of the four-dimension model. Separate assessment of each of the four motivation dimensions with Exploratory factor analysis showed unidimensionality for all dimensions. Exploratory factor analysis provided a best-fit model of three factors (influence of other people, intention of well-being and emotional aspect of motivation). Re-administration of the instruments showed good test-retest reliability in all motivation dimensions. Female gender, higher education and higher PAAQ score were associated to higher internal motivation and lower external motivation scores. Conclusions The Finnish version of the RM4-FM instrument provides valid and reliable scores in assessing motivation for physical activity in a sample of healthy, older Finnish adults. The score distributions suggested shortages in the scale when evaluating internally motivated subjects. The three-factor model of the RM4-FM provided more

favorable structural validity compared to the original version. This work may aid in future translations into other languages and uses of the questionnaire.

Background
Physical activity has a key role promoting well-being and quality of life 1 . Inactivity has been shown to be a major risk factor for several chronic diseases, such as cardiovascular diseases, diabetes mellitus type 2, osteoporosis and colon cancer as well as psychological disorders 25 . The prevalence of such diseases in western countries is high 5,11,12,15,30,34 .
In addition, cardiovascular diseases are a major cause of death 30,34 . Overall, the burden of such diseases for society are substantial measured in costs and healthy years lived 3,38 .
The evidence of the positive effects of physical activity on the risk and, on the other hand, the prognosis of the chronic diseases is indisputable 26 . Despite these positive effects on health, inactivity is still a major issue in public health 2,16 .
On an individual level, a key point of the decision of whether to be physically active is motivation. The Self-determination theory by Deci and Ryan 9 divides motivation into controlled and autonomous aspects. In the field of physical activity, the autonomous aspect of motivation refers to the intention of joy, well-being and achievement of personal goals derived from physical activity, while the controlled aspect refers to the influence and the pressure of others, as well as the guilt of not exercising 9 . Self-determination theory proposes the importance of the motivational factors originating from oneself, rather than external factors, on the determination between physically active or physically inactive lifestyles. Strong intrinsic motivation has been shown to be associated with positive long-term effects of physical activity, as intrinsically motivated individuals continue being active even when the influence of external pressure has stopped 9 32 . Thus, to achieve permanent positive change in one's physical activity habits, it is crucial to enhance the intrinsic motivation aspect.
Lately, the development of interventions on strengthening the motivation for physical activity has been a subject of interest in the field of sport science and medicine 10,29,31,36 . Such interventions have been effective in enhancing internal motivation for physical activity 10,31 . However, it is important to identify the individuals who are at risk for physical inactivity-related diseases and have issues with their motivation to provide targeted interventions to those who benefit most. For this purpose, several questionnaires have been developed 7,21,23 .
The Motivation for Physical Activity (RM4-FM) questionnaire is a self-administered instrument developed to assess the internal and external aspects of physical activity motivation 7 . RM4-FM was developed based on Self-determination theory. RM4-FM has been utilized to examine motivation for physical activity 24 , yet it has not been properly validated for this purpose. In addition, modified versions of RM4-FM have been used in assessing motivation for dieting and quitting smoking 17 .
Validation and understanding of psychometric properties of measurement instruments is indispensable for correct use of these questionnaires 19,20 . Proper validation processes enable researchers to evaluate the usability of the instrument when assessing whether to choose the instrument when initiating a new study. Additionally, proper validation creates the frames for interpreting and assessing the meaning of the results obtained from these instruments.
Hitherto, the psychometric properties of the RM4-FM questionnaire have not been examined properly. In addition, the instrument has not been translated or validated for the Finnish population. The aim of the present study was to translate the RM4-FM instrument into Finnish and examine the psychometric properties of the translated version in a sample of healthy, older adults.

Methods
The study was conducted as part of an exercise intervention study examining the effect of resistance training frequency on strength, functional capacity and health outcomes (NCT02413112). The study contacted two-thousand 65-75-year old people living in the Jyväskylä city region by recruitment mail, who were identified based on age and selected randomly from the Population Register. Those who were willing to participate to the study (n = 454) completed an electronic preliminary registration questionnaire that was used to select individuals to an information meeting (n = 148). The individuals with health issues that may hamper involvement to physical activity (notable overweight with BMI 37 or above, poorly controlled cardiovascular disease or lower extremity injury that may complicate physical training and testing, a need of assistive device, use of medication that influences the neurological or endocrinal functions) or who already involved regularly in physical exercise or sport (over 180 minutes per week) or had other factors which may affect the physical performance and exercise (previous experience of strength training, previous testosterone treatment, smoking) were excluded. After all, 102 suitable individuals signed informed consent to participate and were included to the study. The baseline characteristics of the participants are shown in Table 1. The participants completed the RM4-FM and the Physical Activity Acceptance Questionnaire (PAAQ) questionnaires electronically, as well as a questionnaire on sociodemographic characteristics. The participants were asked to re-complete the RM4-FM questionnaire one month after the initial assessment. Physical Activity Acceptance Questionnaire (PAAQ) 7 The PAAQ is a 10-item questionnaire measuring the subjective psychological and physical discomfort caused by physical activity 4 . All items use a seven-point Likert-scale and the maximum score is 70. Lower scores indicate higher amount of discomfort caused by physical activity and problems with motivation to continue such activity. The PAAQ has been shown to be valid and reliable in measuring psychological factors behind motivation for physical activity 4 .
Translation and cultural adaptation process

Statistical analysis
The sociodemographic and clinical data are presented as means, standard deviations (SD) and 95% confidence intervals (CI) or as counts with percentages. Normality of the RM4-FM and its dimensions score distributions were assessed. Floor and ceiling effects were examined from baseline measurement to assess the scale targeting. If 15% of patients scored minimum or maximum points, the floor or ceiling effect was considered confirmed 18 .
To test cross-cultural validity, the differences in the motivation dimension scores between gender, educational and relationship status subgroups were assessed using independent samples t-test. Pearson correlation coefficients were calculated to assess the association between RM4-FM and sociodemographic characteristics of the participants. Furthermore, convergence with the PAAQ score was examined by calculating Pearson correlation coefficients for each dimension as well as for RAI against the PAAQ score. The coefficient values of 0.3, 0.5, 0.7 and 0.9 were interpreted as low, moderate, strong and very strong correlation, respectively 22 .
To assess the structural validity of RM4-FM, confirmatory factor analysis was conducted.
The presumed factor structure consisting of four motivation dimensions was tested. The Chi-square test of model fit, the root mean square error of approximation (RMSEA), the comparative fit index (CFI) and the standardized root mean square residual (SRMR) were calculated. A non-significant chi-square test value, RMSEA less than 0.05, CFI over 0.95 and SRMR less than 0.80 were interpreted as good fit of the tested model 28 . The unidimensionality of the RM4-FM dimensions were examined with exploratory factor analysis (EFA). In addition, a best-fit factor model was explored with EFA. Varimax rotation with the Kaiser normalization method was used to obtain the most distinguishable factor structure. A factor loading of 0.4 was used as a cutoff value representing that the item typifies the factor sufficiently 27 .
Internal consistency of RM4-FM and its dimensions were assessed using Cronbach's alpha with bootstrapping method of 1000 repetitions to obtain the 95% CIs 6 . Alpha values over 0.7 were considered as acceptable 28 .
Intraclass Correlation Coefficients (ICC) and 95% confidence intervals were calculated to assess test-retest-reliability. ICC over 0.7 was interpreted as sufficient stability, while ICC values under 0.7 represent unacceptable stability 28,33 .

Sample characteristics
A total of 102 eligible participants completed the questionnaires in their entirety. Table 1 presents  Table   2). No other floor or ceiling effects were observed.   Table 2).
Examination of the best-fit factor model of RM4-FM with EFA revealed three factors with eigenvalues over 1 (Table 3) Cronbach's alphas of these factors were 0.86, 0.85 and 0.81 for factors 1, 2 and 3, respectively. Table 3 Factor loadings of RM4-FM items in the best-fit factor model. Test-retest reliability Eighty-one participants (79%) re-completed the RM4-FM. Figure

Discussion
The main finding of this study was that the Finnish version of RM4-FM provides valid and reliable scores in physical activity motivation assessment. Nevertheless, our findings showed limitations regarding the targeting of the dimension scores and a need for consideration of factor structure of RM4-FM (Table 4). There is a lack of validated instruments concerned with motivation of physical activity for the Finnish population. In this study, the translation and cross-cultural validation of the RM4-FM was performed according to internationally submitted guidelines 37 . There were only minor discrepancies between two forward translated versions, as well as between the original and backward translated versions, and these discrepancies did not High ICC values indicated high test-retest reliability with one-month re-completion interval. Motivation for physical activity is a dynamic phenomenon that may vary depending on changing situations and mood of one's life. High ICC suggests that RM4-FM is not strongly affected by these varying circumstances, but it rather reflects more permanent aspects of motivation.
There were some limitations in the present study. Firstly, criterion validity of RM4-FM was not measured, as there was not a sufficient instrument to compare the results. Secondly, the study sample was limited to older individuals (i.e. 65-75 years of age), which, in addition, may be influenced by a high rate of chronic diseases and medication use in the sample. The age and health status of the participants may have also affected habits and attitudes, as well as motivation for physical activity, particularly since the present sample were highly functionable for their age despite the fact that they were not physically activity. The advantages of this study were a large randomly selected sample from the Population Register, which represents well the average population as the regularly physically active individuals were excluded, as well as comprehensive translation and statistical methods, which adhere to the international guidelines on questionnaire validation studies. In addition, the questionnaires were completed electronically, and the questionnaire program required the participant to complete all the items, thus, preventing missing data.
In conclusion, the Finnish version of RM4-FM instrument provides valid and reliable scores in assessing motivation for physical activity in a sample of healthy, older Finnish adults.
However, regarding the floor and ceiling effects of the dimensions, the scale seemed to be improperly targeted suggesting shortages in the scale when evaluating internally motivated subjects. A three-factor model of the RM4-FM provided more favorable psychometric properties than the original version though it loses the basis of Self-Determination Theory. The three-factor model is a noteworthy alternative when assessing motivation for physical activity using the RM4-FM. This work may aid in future translations into other languages and uses of the questionnaire.

Ethics approval and consent to participate
Ethical approval was granted by the Ethics committee of the University of Jyväskylä, Finland and all participants signed informed consent prior to the study initiation.

Consent for publication
Not applicable

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the physical activity and weight control: a randomized controlled trial in women.