5.1 THE INSTRUMENT
This study was conducted in Ethiopia in 2010. The data were collected from Addis Ababa (urban) and rural areas. The quantitative method was used to describe, explain and establish the association between health behaviors and sociodemographic factors among adolescents through a
self-administered questionnaire. Quantitative methodology was used. Thus, quantitative (numerical) data were collected from high school students.
In this study Global School Health Survey (GSHS) questionnaire that of the Ethiopian version was used. The overall objective of GSHS is to increase our understanding and knowledge of health and health-related behaviors, the top leading causes of diseases, disabilities, and death among
adolescent people in their social context (Currie et al. 2000, Aarø et al.
1986). In other words, the Global School Health Survey consists
standardized self-report questionnaire filled out in the classroom aimed at determining and predicting the distribution and the prevalence of an extensive range of health behaviors that determine the health and
wellbeing of adolescents (McMillan et al. 2015, Currie et al. 2009, Freeman et al. 2008). The Ethiopian version of the GSHS questionnaire is prepared in the English language. The translation into the Amharic language was made by the researcher (Jembere Berhanu) and then translation from the
Amharic language to the English language was made by a bilingual person (native Amharic speaker) in both Amharic and English languages.
Additionally, the Ethiopian version of the GSHS questionnaire was translated into the Oromo language by the main researcher then back-translation from the Oromo language into the English language was made by the Oromo language teacher (native Oromo speaker). This was done to verify that the GSHS questionnaire both in English and Amharic and English and Oromo language do have the same meaning and purpose in both languages. Each question was reviewed for semantic and conceptual equivalence in meaning (Behling 2000) again and again to make sure each
item fits into adolescents’ cultural contexts in Ethiopia. Moreover, the GSHS questionnaire was pilot tested on 30 adolescent students who had similar ages with the present study population (15 on Amharic speaking
adolescents and 15 on Oromo language-speaking adolescents) in Ethiopia before it was administered in nominated schools to ensure the accuracy of the questionnaire in adolescents’ cultural context in Ethiopia.
Consequently, following the pilot test of the questionnaire, the GSHS questionnaire was further modified and tailored into Ethiopian adolescents’ cultural context. Global School Health Survey (GSHS) is a previously validated questionnaire or tool and it has been used and
adapting to local culture worldwide (Currie et al. 2000) although no country reported its validity and reliability before.
The Ethiopian-specific GSHS questionnaire consists of 49 multiple-choice questions, including 10 questions on sociodemographic factors (gender, age, grade level, residence areas, living condition, religion, parents’
education, and occupation) and 39 questions on food eating patterns, hygiene behaviors, mental health, substance use, sexual behaviors, physical activity, leisure-time sedentary behavior, and protective factors.
The food eating habits (hungry, fruit, and vegetables) were assessed with three questions or items (Appendix I)
The hygiene behaviors (dental and hand hygiene behaviors) were assessed with three questions (Appendix I).
The mental health (loneliness, sleep patterns, feeling sad or hopeless, bullying, suicidal behaviors, and close friends) was assessed with seven questions or items (Appendix I).
The substance use behaviors (smoking behaviors, consuming alcohol behaviors, drug use behaviors) were assessed with 12 questions or items (Appendix I).
The Sexual behaviors were assessed with five questions or items (Appendix I)
The physical activity (running, walking, biking, playing football) was assessed with four questions or items (Appendix I).
The leisure time sedentary behavior (more than two hours sitting in a day) was assessed with one question or item (Appendix I).
The protective factor (school attendance, perceived adult and peer care and support in school setting, parental regulation, supervision, monitoring, bonding) were assessed with four questions or items (Appendix I).
5.2 DATA COLLECTION PROCESS
A self-managed GSHS questionnaire was used for the collection of quantitative data and it was adjusted to fit Ethiopian school adolescents’
cultural context. The data were collected at two schools from school-going adolescents who lived in rural and urban areas. One school was a typical rural school (West Ethiopia) and another school was a typical urban (Addis Ababa) school. In this study adolescents aged from 11-17 years
participated. Many high schools were contacted with the consultation of the local education bureau and consequently, two high schools showed their willingness to take part in the present study. Teachers were
instructed prior to the data collection. Teacher data administrators were given comprehensive instructions and a list of randomly selected
classrooms from each school. Teacher data administrators and principal researcher introduced the survey and provided orientation to classes several days in advance to data collection. The questionnaires were distributed and collected by these trained and instructed schoolteachers.
Classrooms were selected randomly from a list of classrooms from two high schools from rural and urban areas to obtain the desired sample (by lottery). Overall, 1000 high school students were invited to participate (500 students in rural areas and 500 students in urban areas). All students in selected classrooms were invited to participate except students who were not qualified because of their ages. Parents of participating adolescents were informed prior to the data collection.
Adolescent students were given information in school and asked to take consent letters to their parents to get parent approvals to take part in the present study. In other words, parents or guardians were required to sign and return the consent form for the sake of approving their adolescents' participation in the present study. Written and signed parental and adolescent students’ consents were obtained prior to data collection.
Adolescent students were informed of the purpose and the nature of questions that were asked and how their anonymity and privacy will be masked and protected. Additionally, they could choose to participate or not, that they could skip questions they did not want to answer, and that nonparticipation would not affect their grades in the class. Adolescent students were informed and allowed to decline participation at any time however no adolescent students refused to give their consent for
participation. All adolescent students took part cheerfully and voluntarily in the present study. Adolescent students who did not get the chances
because of the random sampling technique and because of age limits asked why they were not allowed to participate in the “examination.”
Consequently, the researcher and trained teachers for data collection explained to adolescent students the reasons why it is not possible for all students to take part in this study. Trained teachers for the collection of data were advised to avoid circulating the class, instead, to stay or remain at the front of the class during data collection to protect privacy and anonymity. Additionally, students were instructed not to look at or follow their classmates’ responses when they filled out the questionnaires.
Rather, students were encouraged to respond to each item that was regulated and informed by their own perceptions and opinions. The students were asked to give their honest responses to each question because there is no specific correct answer for each question (item). The answer for each item can be different for different students. Every
identifying information of adolescent students was masked in the present study. Each adolescent respondent’s responses were confidential.
5.3 DATA ANALYSES
The characteristics of respondents and health-related behaviors were assessed, described, and explained using cross-tabulation analysis. The findings were presented as frequencies (N) and percentages (%). And the association between each respondent characteristic and the likelihood of health-related behaviors was analyzed using logistic regression model.
Logistic regression models were used to estimate the odds ratios (ORs) of
each health behavior outcome. And confidence intervals (CI) should be provided for each of the study estimates presented throughout reporting findings, providing the likely range of values to be found in the sample population being studied. Logistic regression is a statistical process that produces findings that can be interpreted as an odds ratio (OR)
(Pourhoseingholi et al.2012). Findings were presented as odds ratios (OR) and 95% confidence intervals (CI). And only variables that were statistically significant at p<0.05 levels were included in the final report and variables that were not statistically significant were excluded. The data entry and statistical analyses were carried out using SPSS version 25:0 software.
5.4 ETHICAL ISSUES OF THE STUDY
In the present study, the General Data Protection Regulation (GDPR), which is to restore trust by empowering and putting citizens in control of their data is applied. Thus, information about the present study in a concise, transparent, and understandable form, using clear and plain language was provided to adolescents. Thus, each adolescent was made easily
understand why and how his or her data is processed and used.
Additionally, children are required to have a certain minimum level of understanding and requirement to give consent of what they are being asked and why they are being asked. Thus, GDPR necessitates those parents or guardians to give consent to the personal data processing of young children. Furthermore, this study was done in line with the directive articulated on Helsinki Declaration in 1964 and its modifications.
Therefore, written assent from all the adolescent participants and written informed consent and their parents were obtained. Ethics research approval from Eastern University Hospital Ethics and Review Committee (Pohjois-Savon sairaanhoitopiirin tutkimuseettinen toimikunta-KYS81028-9M 09.09) was granted. The local Education Bureau and schoolmasters have granted permission to conduct this study in Ethiopia. The consent forms for parents and adolescent students were translated into Amharic and Afaan Oromo. The schoolmasters and trained teachers for data collection distributed consent letters to parents through adolescent
students and consent forms to the adolescent students. Then adolescents who returned completed parent consent forms to the school master’s office in each school before the data collection were approved to participate and fill out the self-administered questionnaires.