*2.4. Data Collection*

Data collection was carried out from January to March 2020 by the first author—a Lebanese female doctoral researcher, who is an Arabic native speaker. Data collection was completed using a tablet computer, on which the questionnaire was programmed employing the Magpi® application. Data were collected one-on-one in a private environment, either in the participants' or in the gatekeepers' apartments.

#### *2.5. Data Analysis*

Data were analyzed using IBM SPSS Statistics version 27.0.(International Business Machines Corporation, New York, NY, USA) A descriptive presentation of the results of the questionnaire is given for continuous variables that are non-normally distributed through interquartile range (IQR) and medians. Since none of the variables were normally distributed, standard deviation (SD) and means were not calculated. Tests of associations were conducted for categorical variables using Fisher's exact test, since the study's sample size, and in consequence size of cells, is considered small [42,43]. The Chi-square test was used to check for significant differences between proportions across categories (SRH service categories). A threshold of significance was set at 0.05. No data were missed.

In order to evaluate the overall knowledge of participants on SRH issues, an unweighted score was generated for every participant based on her knowledge on different SRH topics, as reported by Ivanova et al. [44] in a comparable study in Uganda: STIs, symptoms of STIs, methods of contraception, and danger signs of pregnancy. Each of these elements were assessed through a scale from zero to three. After combining the evaluation from the four elements and getting the final average score, the overall knowledge on SRH issues was described as following: low (average score ≤ 1), medium (average score between 1 and 2), and high (average score ≥ 2) [44].
