Aim
The present study aims to adapt RMC to the Iranian culture and determine its psychometric properties.
Study’s participants
The inclusion and exclusion criteria for this study have been published previously in another article [12].
Sample size
Selecting of ten participants per item has been suggested for factor analysis by Nunnally and Bernstein [13]. Because the RMC scale has 15 items, therefore, 150 participants were needed. With cluster sampling and a design effect of 1.5, the sample size was found as 225, which was increased to 265 to take account of a potential withdrawal rate of 20%.
Tool
The detailed information about the tool has been published in the protocol paper [14].
Translation process
First, written permission for adapting the tool to the Iranian culture was obtained from the tool developer (Sheferaw). The original version of the tool was translated from English into Persian by a native English speaker who was also competent in Persian language. The translated version was reviewed by the research team, and then translated back from Persian into English. This step of the translation was carried out by two translators competent in both languages who had not been involved in the forward translation. Next, this translated version was reviewed by two people familiar with specialized concepts and competent in both languages and the final version was thus obtained [15]. The Persian and English versions are available as appendix 1 and 2, respectively.
Data collection
The study was conducted in the postpartum ward of public (Alzahra, Taleghani) and private (Behbood, Nor-e-Nejat, and Shahriyar, 29 Bahman) hospitals in Tabriz. A total of 265 postpartum women were selected. The questionnaire included socio-demographic, obstetrics characteristics, and the RMC scale. The demographic questionnaire used contained questions on the mother’s age, education, occupation, income, the neonate’s gender and pregnancy type (intended or unintended). The validity of this questionnaire was confirmed using content validity.
Face and content validity
To determine the face validity of the scale, 20 postpartum women were invited to assess all the items in terms of simplicity, clarity and relevance. Then, based on their responses and the Likert-type scale (from 1 point = ‘totally difficult or unclear’ to 4 points = ‘totally simple and clear’), the item impact was determined for each item using the following formula: Impact = Importance (mean responses to the item) × Frequency (the number of responses with the score of four). The items scoring less than 1.5 are eliminated [15].
Content validity was determined by both a quantitative and a qualitative method. In the qualitative method, ten experts in reproductive health, midwifery and psychiatric nursing were asked to assess the translation of each item in terms of grammar, use of appropriate terms and correct placement of the items and to present their corrective comments. In the quantitative method, the Content Validity Ratio (CVR) and Content Validity Index (CVI) were measured. To determine the CVI, the items were assessed in terms of relevance, clarity and simplicity using a 4 point Likert scale. Scores above 0.79 were considered acceptable. CVR was determined by experts who asked to evaluate each item in terms of importance using a 4 point Likert scale. The minimum CVR was taken as above 0.62 based on Lawshe’s table.
Construct validity
Exploratory (EFA) and confirmatory (CFA) factor analyses were used to assess the construct validity. Bartlett’s test, the Kaiser-Meyer-Olkin (KMO) index, scree plots and Oblimin rotation were used in EFA. The adequacy of the data for conducting EFA is confirmed based on values above 0.7 [16]. The factors were extracted by Principal Component Analysis (PCA) and varimax rotation, and the number of factors was determined based on Eigen Values (EVs) and scree plots. EV determines what proportion of variance in the total data is explained by one factor. Therefore, higher EVs for any factor increase the proportion of variance explained by that factor [17].
Factor analysis assesses the intra-variable relationships and is used to extract categories of items most related to each other. The items with a factor loading lower than 0.3 were considered as candidates for elimination, and then the research team decided about whether or not to keep the items where they were greater than 0.3 and less than 0.5. Also, the factors’ consistency with the subscales of the original scale was assessed after the extraction of each factor and the items in the factor.
The structure of the extracted factors was assessed using the EFA model and CFA. The indices were used for assessing the exploratory model’s fit. Fit indices and reasonable values of theses indices for CFA were considered as Comparative Fit Index (CFI) ≥0.90, Root Mean Square Error of Approximation (RMSEA) < 0.08, X2/df < 5, Tucker-Lewis Index (TLI) ≥0.95 and also, Comparative Fit Index (CFI), Goodness of Fit Index (GFI), Adjusted Goodness of Fit Index (AGFI), Normed Fit Index (NFI) and Incremental Fit Index (IFI) > 0.9 [18].
Internal consistency and test-retest stability methods were used to assess the reliability of the questionnaire. To assess internal consistency, the Cronbach’s alpha was determined for a sample of 20 mothers, and to examine the test-retest stability, Intraclass Correlation Coefficient (ICC) was calculated for the same group of mothers, who had completed the questionnaire twice, with a two-week interval. Alpha coefficients higher than or equal to 0.06 were considered acceptable. ICC ≤ 0.4 were considered poor to fair agreement; 0.41–0.60 moderate agreement, 0.61–0.80 good agreement and 0.80 excellent agreement [19].