This study aimed to translate and assess the psychometric properties of SCECQ.
A total number of 5 samples per item were selected as the participants (125 samples for the 25-item questionnaire); however, considering the design effect of the cluster sampling method and a 10% loss to follow-up rate, 205 individuals were selected as the sample.
SCECQ was utilized to collect data on mothers’ satisfaction with childbirth classes. This questionnaire was developed by Lee et al. in Chinese language and was translated in English for the purpose of publications. However, there is no information about the Chinese-English translation process they used . In this 25-item instrument, all items are scored on a five-point Likert scale ranging from “not at all satisfied” (score 1) to “very satisfied” (score 5). The thematic domains of the questionnaire include the class structure (questions 1-5), the class process (questions 6-21), and the class outcome (questions 22-25). The questionnaire demonstrated good content validity with an index of 0.88. The Cronbach alpha for the whole scale was 0.89 and for the structure, process and outcome subscales were 0.76, 0.88 and 0.72 respectively . The present study is proposing an Iranian version of the instrument.
The translation process was conducted systematically by applying the forward and back‐translation method , the original version of the instrument was first translated from English into Persian language by a translator, who was a native speaker for both languages. After reviewing the first version of the questionnaire, the Persian version of the instrument was translated into English by other two professional translators who were not involved in the previous stage. The final questionnaire was prepared after review of two independent individuals who were familiar with medical terms and mastered both languages.
The research protocol of this study was published in 2019 . The study was conducted between July 2019 to March 2020 during a 9-months period (one month for questionnaire translation, one month for content and face validity assessment, five months for sampling, and two months for data analysis and manuscript writing). Seven out of 20 healthcare complexes in Tabriz were selected through simple random method. Each complex covered four to five healthcare centers, and sampling was conducted in a total of 29 centers. The required sample size for each complexes was determined using the proportional sampling method. The participants were randomly selected based on quotas determined for each center. The inclusion criteria for the study were being 35 to 37 week pregnant, living in Tabriz city with no age or language restrictions. Women who had not attended any sessions were excluded from the study. After the inclusion of the eligible women, the study objectives and methods were fully explained to them and socio-demographic questionnaire and SCECQ were completed through interviews. The socio-demographic questionnaire was developed by researchers and included the following items: mother’s age, her educational degree, and her job, her spouse’s educational qualifications and job, her household income status, and having a wanted or unwanted pregnancy. The validity of this questionnaire was confirmed through the measurement of its content validity.
Face and content validity
To assess the face validity of the questionnaire, 20 randomly selected pregnant women were asked to rate the difficulty, relevance, and ambiguity of all questions. The responses were scored on a four-point Likert scale ranging from 1 (completely difficult/irrelevant/ambiguous) to 4 (completely simple/relevant/unambiguous). The respective impact score for each item was calculated through multiplying the mean score of each item (importance) to the number of responses (frequency). An item was considered acceptable, if the impact score for that item was more than 1.5 .
Content validity was assessed through qualitative and quantitative methods. In the qualitative phase, 10 midwifery and reproductive health specialists were asked to review the translated questionnaire and provide feedback on the correct grammar, vocabulary, and phrases in each sentence. CVI and CVR were used in the quantitative phase. CVI values were calculated through determining the simplicity, relevance, and unambiguity of the items and scoring them on a four-point Likert scale. A CVI value of higher than 0.79 was considered acceptable. To determine CVR scores, the experts were asked to comment on the necessity of each item by using a four-point scale. Based on the Lawshe Table, the minimum acceptable CVR value was determined as 0.62.
A scale-based EFA and a CFA were performed to assess the construct validity of the questionnaire. Factor analyses were also performed for the items at the subscale level as well as the whole questionnaire.
Exploratory Factor Analysis
EFA was performed through the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test of sphericity. Values above 0.7 confirm the adequacy of the data for conducting EFA . In addition, eigenvalues and scree plot were utilized to determine the number of factors. An eigenvalue is a measure that determines the amount of variance in a dataset explained by a factor; therefore, factors with higher eigenvalues explain more variance .
Factor analysis assesses inner-variable relations and is used to extract a group of items that are most closely related to each other. In this analysis, items with factor loadings of < 0.3 were omitted, and the research team decided whether to accept or omit those with factor loadings between 0.3 and 0.5 .
Confirmatory factor analysis
To assess the structure of factors obtained from the exploratory factor analysis, the model was fitted using the confirmatory factor analysis. This factor analysis investigates the confirmation of the exploratory model theoretically and the relationship between factors. The fitness of indices was used to evaluate the model fitness. The following indicators were considered to confirm the acceptable model: Root Mean Square Error of Approximation (RMSEA) <0.08, Standardized Root Mean Square Error of Approximation (SRMSEA) < 0.08, Comparative Fit Index (CFI) ≥ 0.90, Tucker- Lewis Index (TLI) ≥ 0.95, Normed Chi-square (x2/ df) < 5.0.
To determine the overall reliability of the questionnaire, internal consistency and test-retest reliability were measured. Internal consistency was assessed by calculating Cronbach’s alpha in a sample of 20 mothers. Based on rule of Thum, the sample size for reliability is about 10-20% of the total sample size . The test-retest reliability was also assessed by calculating ICC for the same participants who completed the questionnaire twice at a two-week interval.
This study was taken from a PhD thesis in midwifery approved by the Ethics Committee of Tabriz University of Medical Sciences (Ethics code: IR.TBZMED.REC.1398.066). Informed written consents were obtained from all the participants.