Study design and period:
This cross-sectional study was performed in 2015 in Iran
The cluster sampling method was used for sampling. The country's provinces were divided into three provincial clusters of 9, 9, and 13 as follows, and one province was selected from each cluster. Since the sampling was performed with a combination of systematic classification, cluster, and random methods, to reduce the error, the sample size selected from each cluster was proportional to the volume of that cluster.
Multi-stage sampling was performed as follows:
Step 1: the provinces of the country divided into three clusters.
Step 2: from cluster number 1, Kurdistan province; From cluster No. 2, Central; And randomly selected from cluster number three of Fars province.
Step 3: Marivan city from Kurdistan province, Saveh city from Markazi province, and Gerash city from Fars province were randomly selected.
Step 4: to conduct the study and cooperation with the questioners, the Vice Chancellor for Research and Technology of Kurdistan University of Medical Sciences corresponded with the education of all the provinces of Kurdistan, Markazi, and Fars.
Step 5: The list of primary schools in the cities of Marivan, Saveh, and Gerash obtained from the same city's training. Then, six primary schools (three girls 'primary schools and three boys' primary schools) were randomly selected from primary schools in each city.
Step 6: From the selected schools of each city, there were 526 samples (263 girls students and 263 boys students) selected from 10 to 12-year-old children (grades 5 and 6).
The study sample included 1590 boys and girl children with the age of 10-12 years old.
Based on the list received from schools, the volume of the allocated sample selected as a multi-stage study of 10 to 12-year-old children (fifth and sixth grades) in each province.
To collect the required data. The children answered one part of the questionnaire, and then they took home the questionnaire, and their parents answered the other part of it at home, and the next day, they brought back the questionnaire. The Demographic variables, including height, weight, and calculating a person's body mass index or BMI for children, were measured by trained public health professionals.
To determine the socio-economic situation in this study, we used a questionnaire that has been presented by O’Donnell et al.
These questionnaires evaluated various variables such as education, the job of the head of the household, different house stuff, etc. Based on the Principal Component Analysis (PCA) method, the variables that had the most significant impact on the variance of all variables were identified at first. Then a new variable (SES) was created based on these variables. According to this index, the population was divided into five quintiles of very poor, poor, moderate, rich, and very rich (14-18).
The individual body mass index (BMI) (weight in kilograms divided by height in meters) was calculated. According to the WHO, a set of sexual BMIs was used to define obesity and overweight, which is a standard for various studies of obesity in children and adolescents aged 2-29 years worldwide. Accordingly, overweight is defined as between 85% and 95%, obesity is above 95%, and slimming is below 5% (19).
In addition, screen time and phone or tablet use were evaluated by asking them how much time they spent each day watching TV or playing with the computer, using smartphones or tablets (for playing the game, talking, texting, or some other use). Spending more than two hours per day was considered high screen time, phone, and tablet use (20).
In this study, a total of 1590 questionnaires were completed, and each questionnaire given a numeric code between 1 and 1590. The data were entered into SPSS software version 20 by two trained experts.
In this study, curve concentration method, Index, Concentration index, and Odds Ratio were used to measure inequality.
After the inequalities were measured, the next step was to decompose them. In the Oaxaca Decomposition Oaxaca method, the difference between the mean of a result in two different groups is explained by the determinants of socioeconomic inequality (determinants) and the magnitude of their effect on inequality. Descriptive analysis of data was evaluated using mean, relative, and mean descriptive indicators. Each of the indicators of Screen Time, activity and phone & tablet use levels was considered as the response variable. To analyze the level of physical activity, Screen Time, and using phone & tablet, the Chi-Square test was used to estimate the prevalence of variable response in each level of demographic variables and then multivariate logistic regression was used to estimate the final model based on variables with p <0.1 in Chi-Square test and single variable OR and AOR calculation. All analyzes were performed with Stata / SE 14.0 software.