This was a methodological study conducted in Tehran, Iran, from May 2014 to September 2015.
The SRH-IQ contains eight domains including: source of information on sexual and reproductive health (16 items), sexual and reproductive health knowledge (26 items), sexual ideology and attitude (24 items), protective behaviours (31 items), contraceptives (28 items), current sexual partners/encounters (35 items), sexual and reproductive health services (17 items) and sexual and reproductive health outcomes (16 items). Of these eight domains, three domains: sexual and reproductive health knowledge, sexual attitude, and sexual permissiveness are measured by a three-point Liker scale (1- agree, 2- not sure, 3-disagree or 1- correct, 2- not sure, 3- false), and the remaining domains are measured based on yes/no format or descriptive categories.
The SRH knowledge contains 26 items in four sections: knowledge of physiology (3 items), sexually transmitted infections/ HIV (7 items), contraception (11 items) and specifically condoms (5 items). A 2-point scale was considered for each question. A score of 2 was assigned for each correct answer, while a score of 1 was assigned for an uncertain or wrong answer.
Sexual attitude includes 23 items on attitudes toward sexual permissiveness (9 items), coercion (2 items), abortion (2 items), peer influence (2 items) and condom (8 items). A score of 3 was assigned for each “liberal” (higher sexual permissiveness) answer, while a score of 2 was assigned for an uncertain and a score of 1 for “conservative” (lower sexual permissiveness) answer.
After securing permission from the developers of the questionnaire, Persian translation was carried out simultaneously by two linguists (English – Persian translation experts) using the method recommended by the World Health Organization (WHO) (18). Two translations were composed by the experts to create a single translation. Then, the questionnaire was translated back into English by two translators, one English speaking fluent in Persian and another Persian speaking fluent in English. To confirm the conceptual uniformity and synonymy of the words and phrases, the English version was compared with the original version by other English language and sexual and reproductive health experts and an agreed upon final version was adapted. With a careful review of this version by an expert in instrument designing and three experts in sexual and reproductive health, themes and phrases that were irrelevant to the Iranian society were modified or deleted. Finally, the provisional version of the Persian questionnaire was developed.
Face and Content validity
Qualitative and quantitative methods were used to assess the content validity. For qualitative methods, a multi-disciplinary panel of experts; including SRH specialists, public health professionals, psychiatrists, epidemiologists, sociologists and anthropologists reviewed the instrument in detail. They paid attention to Persian grammar and syntax, using appropriate words to fit the selected meaning and eliminated unnecessary items, which were considered culturally inappropriate. The experts also made the decision to organize the questionnaire, so that it could be self-administered and reformulated the format accordingly. The University Ethics Committee also suggested a few questions be rephrased to be more ethically sensitive and viable in the best interest of the people answering the questions.
In the quantitative method, the content validity ratio (CVR) and the content validity index (CVI) were measured.
A limited number of questions needed to be modified in terms of clarity and transparency and subsequent modifications were sent to the experts and approved. To assess the CVR, academics in the field of SRH were brought on as advisors to rate each item as “necessary”, “helpful but not necessary” or “not necessary”. Eight out of 10 professionals completed the task, and the answers were calculated according to the Lawshe formula (19).
To assess the CVI, the Persian version of the questionnaire was presented to 14 SRH specialists (different from previous group) and they were asked to evaluate each item in terms of "simplicity and fluency", "relevance" and "clarity and transparency" on a four Likert scales, then the CVI was calculated for each item (I-CVI). Eight of the experts completed the task. Based on the mean scores of the content validity index of the questionnaire items, the average content validity index of the questionnaire (S-CVI/ Average) was calculated. Also, the scale level content validity/ Universal agreement (S-CVI/UA) of the questionnaire was calculated. This indicator expresses the ratio of the total phrases that all experts have scored 3 or 4 (20).
To obtain the face validity, the instrument was piloted with ten people of the target group from different regions of Tehran to assess it in terms of simplicity, clarity and readability. Items were reviewed and amended according to the pilot group’s comments.
Phase 2: Construct validity and Reliability
The sample size was determined according to the number of items (21). Therefore, a purposeful sample of 800 males and females (aged 15–49 years) residing in Tehran, the capital of Iran, were recruited using convenience sampling method in June 2014. Trained staff explained the objectives of the study to the individuals who agreed to complete the questionnaire. Written informed consent was taken prior to providing the self-administered questionnaire.
Female staff assisted female participants to complete the questionnaires and male staff assisted the male participants, in a culturally sensitive manner, and the participants were ensured that they are free to withdraw from the study at any point without any judgment, stigma. The issues of confidentiality as well as de-identifying data procedures that were later carried out were also explained to potential participants.
Inclusion criteria consisted of: a certain level of literacy that led to their ability to understand and communicate in Persian, residence of Tehran, not having any physical or mental debilitating conditions preventing replying reliably to the questionnaire, and being in the age range of 15 to 49 years.
To determine the construct validity of the Persian version of the instrument, exploratory and confirmatory factor analyses were performed. We used LISREL 8.8 software for confirmatory factor analysis (CFA), and SPSS version 16 for exploratory factor analysis includes Bartlett specificity test and Kaiser-Meyer-Olkin (KMO) for assessing the proportionality of the factor analysis model and adequacy of sampling respectively. We used the principle components analysis with method of Varimax rotation (22).
In the present study, minimum load factor was considered 0.3 for each item. After extracting the factors and expressions in each factor, the consistency of the factors with the concept and the main aspects of SRH was investigated.
To examine model fitness, the following goodness-of-fit indices were used: the adjusted root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), Tucker-Lewis index and comparative fit index (CFI) (23). The cut-off values of ≤ 0.06, ≤ 0.08, ≥ 0.95, were thought acceptable for RMSEA, SRMR, TLI and CFI respectively (24).
The Cronbach’s alpha was used to ensure the internal consistency of the measurement. The Cronbach’s alpha higher than 0.6 was considered acceptable (25). Test-retest analysis was used to determine the stability of the questionnaire. A sample of 60 individuals (30 men and 30 women) completed the questionnaire twice with a two-week interval.
Stability for binary items was assessed using the Kappa statistics (26), and for Likert scales using the intraclass correlation coefficient (ICC) (27).
All the processes of this study were approved by the Ethics Committee in the Isfahan University of Medical Sciences (reference: 393460).