Received date: June 22, 2017; Accepted date: July 07, 2017; Published date: July 20, 2017
Citation: Darabi F, Farahani FK, Yaseri M (2017) Psychometric Analysis of the HIV Behaviors Questionnaire for Female Adolescents (HBQFA) in Iran. J Psychiatry 20: 421. doi: 10.4172/2378-5756.1000421
Copyright: © 2017 Darabi F, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Psychiatry
Background: In the absence of a valid instrument for measuring theory-based HIV/AIDS behaviors in highschool adolescent girls in Iranian. This study was aimed to develop and evaluate the psychometric properties of a questionnaire to assess HIV/AIDS related behaviors among female adolescents. Materials and methods: In this cross-sectional study 578 female adolescent participants were enrolled. Multistage cluster randomization sampling techniques was used to select study participants. The data on demographic and components of the TPB (theory of planed behavior) model were collected through self-maid using structure questionnaire. Content, face, and construct validity analysis of the questionnaire were assessed. Results: Results showed that the data was fit to the model (χ2=39222.95, P<0.001). Exploratory Factor Analysis (EFA) (KMO=0.73) with varimax rotation was used to construct validity. For optimal reduced solution 18 items and 6 factors were used. These factors jointly accounted for 63% of the observed variance of outcome variable. The confirmatory factor analysis indicated good fit to the data (RMSE= 0.045 95% CI 0.038 - 0.052). In addition, Cronbach’s alpha coefficient was showed an excellent internal consistency (alpha=0.94). Conclusion: This study finding verified that, the factor structure of the expanded TPB scale of HIV of adolescent’s. Providing and applying valid and reliable questionnaires are necessary to assess HIV behaviors that need intended intervention.
Psychometric; Questionnaire; Reliability; Validity; HIV; Adolescent
Human immunodeficiency virus related behavior needs of young people in Iran are not well studied. This is concerning at a time when all communities in the world are threatening by morbidity and mortality due to the spread of the acquired immune deficiency syndrome (AIDS) . Based on available evidence 60 million people are infected by HIV globally and 20 million are expected to die due to complications of the disease .
HIV/AIDS epidemic in Iran is growing significantly in past twenty years . According to the last statistics in relation to HIV/AIDS published by ministry of health and medical education, from the start of epidemic to the end of first half of 2014, 28663 people with HIV/ AIDS were identified, and male population consists of 88.4% in Iran . However, there is a belief that this data underestimated the true number of HIV/AIDS cases in Iran, and it is likely to be four times higher than those the published data. Evidence shows that 85% of HIV infected people in Iran are acquired the virus through Drug use and 10% are through sexual contact . Majority of youth are initiating their sexual activity during their adolescence years. Evidence shows that half of new HIV infected cases and one-third of the 340 million new sexually transmitted infections (STIs) occur in people aged under 25 years . According to cultural norm in Iran majority of adolescents are emphasizing on the abstinence until permanent marriage. However, sexual activity before and out of marriage can be happen in both Islamic and non-Islamic societies . Therefore, adopting safe behaviors is essential in order to develop preventive interventions .
Young people comprise a significant proportion of the total population in Iran. In recent national census, there are approximately 12 million adolescents (aged 10-19), and 21 million young people (aged 10-24) which consist nearly less than one third of the population of the country . According to previous study, girls compared to boys are suffering from lack of HIV/AIDS and sexual related knowledge and information . Lack of education, misinformation, embarrassment, hesitation to get into a public created social discussion and psychological obstacles are the main barriers of adolescents reproductive health information . Thus, awareness creation through designing different HIV program among adolescents and their family is essential to prevent the transmission of the virus. Beside awareness creation, preparation and validation of HIV related behavior assessment tool is important to assess behavior change.
Based on other study results adolescents' interest and curious to high risk behaviors are a complex and ambiguous process. There are numerous determinant factors that effect on the beginning high risk behaviors include lock of education parents, the pressure of friends and classmates alongside cognitive factor like attitudes, social norms, and self-efficacy which they have a major relation for high risk behaviors adolescence . However, further studies are needed to determine how to preventive adolescence of high risk behaviors based on evidence will be effective when theory-based approaches such as social cognitive models . Based on evidence the TPB toward other behavior theories more is known as the applicable for explain how adolescents engage in the high risk behaviors.
To predict HIV/AIDS related behavior among adolescent females, TPB was used. According to this theory, there are three determinants of behavior, such as attitudes, subjective norms and perceived behavioral control (Figure 1). These predictors can either affect or predict people’s intention to act out a specific behavior . Some studies have shown that, perceived parental control may improve children’s behaviors . Thus, in this study perceived parental control construct was added to the original theory to strengthen the HIV/AIDS related behavior prediction power of the theory. Several evidences emphasized on the effectiveness of the TPB constructs for high risk behaviors among youth [10-15].
Moreover, there are few studies  have been reported on the assessments of direct measures of the TPB by adding perceived parental control to predict HIV/AIDS related behaviors among adolescents . As Ajzen emphasized on Necessity to design an appropriate questionnaire in relation with risky behaviors . In Iran, there are limited scale are available to assess HIV/AIDS related behavior. For instance, Information-Motivation-Skill-Behavior(IMSB) questionnaire is one of the scale available in Iran to assess HIV-related behavior . Majority of questionnaires used in Iran to assess HIV related behavior are self-made and intended to explore the knowledge and attitudes of students [4,19-22]. However, the validity of the questionnaires and application of the theory are not demonstrated well. Therefore, developing and validation of instrument based on behavioral change theories of social and behavioral sciences are important to use uniform questionnaire to enable comparison between different studies finding. Thus, the aim of this study was to develop and validate a theoretical model based questionnaire to measure HIV/AIDS related behaviors among high schools adolescent students.
Participants and procedures
A cross-sectional study was conducted among 578 female middle school students. The participants were recruited from 6 female schools in three districts in Tehran, Iran. The schools were randomly selected from 36 female middle schools in three districts of Tehran (districts 2, 4 and 10). The schools were comprised a mixture of working-class and lower-middle-class families. From each school, three classes were selected randomly (one class from each grade). Accordingly, a total of 18 classes were selected.
Accordingly unit of randomization was each classroom (clusters). Randomization was performed separatel, a total of 18 classes were enrolled. In each class, 16 subjects were randomly recruited. Randomization was based on the block permuted stratified cluster randomization method. The y in each district and all the class of the schools were allocated randomly to the experiment and control groups. The block size was 4 or 2 and 10. The inclusion criteria for participation in this study were: 1) girls’ adolescence in the 12-16 age range, 2) Living in Tehran, 3) confirmed satisfaction with participating in the study. The exclusion criteria were 1) being absent for more than one session.
The scale development process
A self-maid questionnaire was developed based on TPB theory constricts. Some parts of this instrument were based on the questionnaire developed by World Health Organization . The WHO questionnaire consists of questions about HIV related knowledge and its sources, sexual attitudes and behavior, sexual and HIV services and outcome of HIV/AIDS infection . In addition, other questions were developed based on literature review and qualitative study (eight focus-group discussions with 40 participants) and were added to the final instrument.
The final questionnaire was consisted of 42 questions, and includes socio-economic status (SES) (13 items), knowledge (11 items), attitude (5 items), subjective norm (3 items), behavioral intention (3 items), perceived behavioral control (3 items), perceived parental control (2 items), and general behavior (2 items).
Socio-economic characteristics: Thirteen items were included in the questionnaire to elicit personal information on age, place of residence, family income, family size, family education and job.
HIV knowledge: Eleven items were used to assess HIV related knowledge. The response of each question was scored using 3 categories, such as true, false, and do not know. Each correct answer was given score one, and wrong and “do not know” responses were scored zero.
Attitude towards HIV behavior: Five items regarding attitude towards HIV, derived from relevant literature, were employed. They were scored using a 5-point Likert scale ranged from “strongly agree” with maximum score and “strongly disagree” lower score.
Subjective norms of HIV behavior: Three items were used to assess the influence of important people (parents, friends, and teachers) on their opinion about HIV related behavior. The response was recorded by five points Likert scale ranged from “strongly agree” with maximum score and “strongly disagree” lower score.
Perceived behavioral control (PBC) of HIV behavior: Three items were used to assess the students’ perceptions of behavioral control about behaving in a way that preserves their HIV prevention behavior. Responses were recorded on a 5-point Likert differential scale ranging from 1 (very difficult) to 5 (very easy). PBC questions were designed according to the available literature [12,25].
Perceived parental control over HIV related behavior: Two items were used to assess perceived behavior control. Participants were asked to indicate the extent of their agreement with the items (e.g., “My parents tell me how I can prevent HIV infection”).
Behavioral intention to HIV related behavior: This is concerned with attitude towards the behavior, subjective norm and perceived behavioral control . Three items which ask about attitude towards HIV/AIDS prevention were used. The response of each questions was recorded by five points Likert scale ranged from strongly disagree to strongly agree.
HIV related behaviors: Two items were used to evaluate HIV/ AIDS related skills. Participants were asked to indicate what they might do in a situation related to HIV/AIDS. The response of each item was recorded with 5 options (For example, “I am sure I can avoid risky behaviors such as unprotected sex and tattooing that is linked with HIV transmission). The maximum total point each participant scored was 100. The higher the score indicates the more agreement towards HIV prevention behavior. Then psychometric properties of the theorybased HIV/AIDS questionnaire were performed to evaluate validity and reliability of questionnaire.
Face validity: To determine the face validity of the questionnaire, was applied in two method qualitative and quantitative the questions need to be logical and match with the characteristics of the respondent. For Qualitative phase, ten adolescent girls were recruited using convenience sampling to assess the ambiguity, relevance, and difficulty of each item. At this stage, none of the items was removed. However, 3 items were modified based on participant’s suggestion. Also, quantitative phase was performed the impact score. Items were considered appropriate if they had an impact score equal to or greater than 1.5.
Content validity: Qualitative and quantitative approaches were applied to test the content validity of the questionnaire. First, 10 specialists in health education and health promotion, public health, and HIV areas reviewed the questionnaire to check its grammar, wording, item allocation and scaling. In order to calculate the CVR 13 other specialists were asked to assess each item on a 3-point Likert scale (1= essential, 2= useful but not essential, 3= not essential) during the quantitative stage. Based on Lawshe’s table  items that scored greater than or equal to 0.54 were kept in the scale. As a result, 2 items were removed from the instrument. In order to calculate the CVI, 10 additional expert panelists were asked to determine the relevance, clarity, and simplicity of each item using a 4-point Likert scale. In addition, two separate groups of experts were used for more precision. However, accordance to Waltz and Baussel  items with CVI value greater than or equal to 0.79 were accepted, and 17 items that did not meet this criterion were removed. Eighteen items had a CVI value of greater than or equal to 0.79 were retained.
Construct validity: This validity was investigated using exploratory factor analysis with varimax rotation. The Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test of Sphericity were used to assess the appropriateness of the sample for the factor analysis. Eigenvalues above 1 and scree plot were used to determine the number of factors. Factor loadings equal to or greater than 0.3 were considered as appropriate . Factor analysis was performed to assess the model fitness. Various fit indices such as relative Chi-square (χ2/df), Comparative Fit Index (CFI), Normed Fit Index (NFI), Incremental Fit Index (IFI), Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual(SRMR) were used . Data were analyzed using SPSS version 23.0 as well as AMOS 23.
Reliability: Also, internal consistency reliability was performed for the direct measures of TPB constructs. Internal consistency was evaluated by Cronbach’s α coefficient. Cronbach’s α coefficient of 0.7 or above was considered satisfactory.
Procedure and ethical considerations
This study was approved by Ethics Committee of Tehran University of Medical Sciences with the number 29226. This study also was registered in the Iranian Registry of Clinical Trials with the ID: IRCT2015070623089N2. In this study, for ethical considerations, the participants were informed about the objective and nature of the study, and each participant provided her written consent in her native language (Persian) prior to the study. Also, we were committed to keeping all of the participants’ information confidential.
A total of 578 students participated in the study. The mean age of the participants was 14.1 ± 1.0 years. Almost majority of the participants (46.5%) were the second child in the family. 43.9% of the participants have economic situation average. Father's education and mother education of more than half of the participants respectively (63.8%) and (63.8%) was free Under Diploma and majority of the participants (73.5%) had housewife mothers. A number of students 517 (89.4%) of participants reported that their Mothers were as the source of receiving information (Table 1).
|Age||Mean ± SD||14.1 ± 1.0|
|Order of birth||1||269 (46.5%)|
|Economic situation||Very good||35 (6.1%)|
|Father education||<6||57 (9.9%)|
|Mother education||<6||72 (12.5%)|
|Father job||Working||544 (94.1%)|
|Mother job||Working||153 (26.5%)|
|Primary information source||Mother||517 (89.4%)|
|Other family||15 (2.6%)|
|Paired group||17 (2.9%)|
|Teacher /Advisor||4 (0.7%)|
Table 1: Socio-economic and demographic characteristics of female adolescents (n=578).
In the qualitative face validity, participants stated that the items were readable and understandable. And based on the result of quantitative face validity affects score was greater than 1.5 for all items.
In the quantitative content validity assessment, 3 items with CVR less than 0.62 and CVI less than 0.62 and 0.80 were omitted. The mean content validity ratio, and the mean content validity index (CVI) were 0.64, and 0.74 respectively. In qualitative content validity evaluation of the measure, all criteria; including grammar, wording, scaling of the questionnaire, and item allocation were found to be appropriate.
EFA: In factor analysis KMO ≤ 0.73 and Bartlett’s test was significant (p=0.0000). The sub-groups analysis was shown an acceptable homogeneity level. Principal component analysis with varimax rotation identified six factors with eigenvalues greater than 1 and factor loading equal to or greater than 0.4. These account for 63% of variance observed. The factor loadings were as follows: factor 1(subjective norm) 3 items, factor 2 (perceived behavioral control) 3 items, factor 3 with 2 items including items 15-16 (perceived parental control), Factor 4 (AIDS behavior) 2 items, factor 5 (behavioral intention) 8 items and factor 6 (attitude towards AIDS issue) 21 items (Tables 2 and 3).
|α (n=578)||Mean ± SD||Number of items||Subscale|
|0.84||34.3 ± 26.8||11||HIV Knowledge|
|0.85||48.0 ± 9.4||5||Attitudes towards HIV|
|0.89||33.0 ± 20.2||3||Subjective Norm|
|0.8||33.4 ± 12.8||3||Behavioral Intention|
|0.94||45.1 ± 15.1||2||Perceived parental control|
|0.81||34.2 ± 15.1||3||Perceived behavioral control|
|0.89||42.9 ± 16.1||2||Behavior|
|0.94||40.6 ± 7.9||29||Total|
Table 2: Number of items, and α of Iranian students' AIDS questionnaire (HBQFA) constructs (n=578).
|1. People around me think that I should avoid risky behaviors (unprotected sex, injection of contaminated blood, and tattoos which result in AIDS.||0.833|
|2. Most people who are important to me want me to avoid risky behaviors which lead to HIV infection.||0.823|
|3. Most people who are important to me expect me to avoid risky behaviors that lead to HIV infection.||0.820|
|4. I am confident that I can stay away from contaminated objects such as syringes and needles infected with HIV/AIDS virus.||0.895|
|5. I'm sure that I can avoid risky behaviors such as unprotected sex and tattoos, which transmit AIDS.||0.818|
|6. I can behave normally with people who are HIV-positive.||0.767|
|7. My parents give me enough training and guidance on the subject of AIDS.||0.895|
|8. My parents determine how much I should read on the subject of AIDS.||0.878|
|9. I avoid having contact with sharp objects such as HIV contaminated syringes, and needles infected with HIV.||0.814|
|10. I avoid doing risky behaviors such as unprotected sex and tattoo which transmit HIV.||0.773|
|11. Within next three months, I'm going to avoid risky behaviors (such as transfusions of infected blood, unprotected sex, etc.) that lead to HIV infection.||0.836|
|12. I have planned to avoid risky behaviors(such as transfusions of infected blood, unprotected sex, etc.) that lead to HIV infection||0.808|
|13. I'm going to in the future Your talking about problems such as sexuality and AIDS with parents and family members to share||0.809|
|14. In my opinion, AIDS is a serious problem for the health of all people.||0.811|
|15. In my opinion, people with AIDS should inform others about their condition.||0.755|
|16. If my friend gets AIDS, I will cut my relationship with her.||0.732|
|17. If a family member gets AIDS, he/she should be left alone.||0.730|
|18. People with AIDS should be kept away from school.||0.721|
|Explained Variance (%)||22.93||10.83||9.35||8.48||6.61||5.66|
|Cumulative Variance (%)||22.93||33.76||43.12||51.61||58.23||63.89|
Note: Factors: 1=Perceived behavioral control, 2=Subjective Norm, 3=Attitude, 4=Behavioral Intention, 5=Behavior, 6=Perceived parental control
Table 3: The results obtained from EFA with varimax rotation among adolescents aged 12–16 (n=578).
CFA: The results of the CFA of the general model with 18 items in six subscales showed that the model was accepted in its current form (the relative chi-square (x2/df) =1.95<3, P<0.001; RMSEA=0.058>0.08, (95% CI=0.050-0.074); CFI=0.953 >0.9; IFI=0.955>0.9; TLI=0.945>0.9; GFI=0.919>0.9; AGFI=0.936). Therefore, the CFA shows the adequacy of the model and the proper fit of its structural model for the study population (Table 4 and Figure 2).
|Summary of rules of thumb||<0.5*, <0.1**||0||>0.8||>0.9||<0.6, <4|
Note: Df: Degrees of Freedom; SB Chi-Square: Satorra–Bentler chi-square; Prob: Probability
* Good fit; ** Mediocre fit
Table 4: Goodness of fit indexes for AIDS dimensions (n=578).
Reliability: The overall and domain specific items were shown excellent internally consistency with Cronbach’s (alpha) coefficient 0.94 for the whole scale and ranged from 0.81 to 0.94 for domain specific (Table 2).
Theory of planed behavior provides an important theoretical structure in relation with the complexities of adolescents’ social behavior. The measurement of theoretical constructs is one of the most important the parts in the study of theory-based health education. In this study the measurement of the TPB constructs was performed by direct method which the general attitude of people is measured towards certain behaviors . TPB theory emphasize on three constructs include attitude, subjective norm and perceived behavioral control. In interventional studies through comprehending, cognitive beliefs, affecting on elements that cause encouraging and motivating people to perform certain behaviors are identified. In our study, the constructs of the TPB were evaluated by direct method .
Validity is one of important attributes of the instrument. The aim of validity is the ability of an instrument to measure what it has been designed to measure. The most important step in determining the validity of a questionnaire is construct validation, especially in the social, cognitive and psychometric issues. That the best method in ration with this topic is exploratory factor analysis . In this study we used a construct validity approach HIV/AIDS related HBQFA questionnaire and that is one of the main strengths of our study. Results from exploratory and factor analyses indicated good structure. In the beginning of plan the questionnaire included of 25 questions. That after the implementation EFA removed 7 Questions from the original questionnaire. In final 18-item questionnaire was classified into six subscales. EFA with Varimax rotation indicated that six subscales including attitude, subjective norm, perceived behavioral control, perceived parental control, behavioral intention a behavior could be extracted. It seems that a careful selection of items related to HIV related behaviors might be the reason why we obtained such satisfactory results. In this respect, based on believe of Ajzen, must be consisted any construct of at least 3 items that by considering the fact, at least 12 items in direct measurement method . In our questionnaire in each segment, there was at least 3 items.
The results of this study in this regard with the theoretical foundation and background of the TPB is consistent. Also, the results of the analysis due to the KMO index on this issue emphasize that sufficient sample size and favorable factor analysis. In this study, KMO statistic was confirmed. Six factors identified in this study explained 64% of the variance and the highest expressed changes were related to the perceived power, this results with Karimi et al.  study results is consistence, that the theory explained 61% of variance of tobacco use in youth, also, in Ghazanfari et al.  study the TPB explained 62% of the variance of physical activity. That in both of these studies attitude had the highest amount of explanation of variance. Similarly, Confirmatory factor analysis (CFA) was shown appropriate factor structure scale. The data show normative patterns in this study are similar with previous research and dimensions questionnaire is a good fit [34,35]. The AIDS/ HIV scale with 18-items was used in this study and had good size and so could determine AIDS/HIV in the six components.This provided comprehensive information on Atittude, subjective norm, behavioral adolescents in relation to AIDS prevention. Eskandari  and Kelly  study are consistent with our results in this regard. In a survey of CVI and CVR was obtained a reasonable value. In Eskandari et al. study  in relation to assess the reliability and validity of international AIDS questionnaire for Iranian student for CVR and CVI values are higher than 0.7 were approved. Also, with Alimohammadi et al.  study was consistence.
The reliability is indicative accuracy and stability of the questionnaire. Findings from determining the reliability showed that Subscales and total scale had acceptable internal consistency coefficient. Results Cronbach's alpha coefficients between 0.81 and 0.94 for all subscales, that this implies that questionnaire had acceptable reliability. Internal consistency reliability assessment resulted in Cronbach’s alpha >80 for all dimensions and total scores, suggesting the items in the HBQFA are homogeneous within the six HBQFA domains. Which is consistent with the Bordewich et al.  study. Also the internal consistency of the TPB in five European countries showed from 0.52 to 0.89, Diamond  study on adolescents, too was reported internal consistency 0.76 for the constructs of the TPB. That corresponded to the view of Ajzen .
Test-retest reliability results among the participants indicated ICC ≥ 0.80 for all HBQFA domains and total scores suggesting the HBQFA is consistent at measuring AIDS/HIV over time. This study is consistent with previous studies, Included in the Kelly  study, confirmed the validity of the six-factor structure for questionnaire HIV Disability Questionnaire with adults living with HIV in Ontario, results reported by Hales et al. ; and Hajian Motlaq  was reported at Cronbach’s alpha coefficient of 0.91. It seems that increasing the number of items can increase the reliability of the questionnaire, therefore necessary in future studies is recommended. Finally, we conclude that the HBQFA is a multidimensional, valid and reliable instrument to measure HIV/ AIDS behaviors among students. Furthermore, according to the result of this study, HBQFA with 18 questions in 6 factors can be used to assess the HIV/AIDS behaviors among students in Iran.
This instrument was shown a number of strengths. The importance of our questionnaire was based theory that in relation to AIDS had not been prepared for students. Also, the advantage of the current instrument over the previous instrument was evaluation of psychometric aspects of the behavior, while the previous instruments only examined domains of knowledge and attitude. Without losing any important dimension on HIV related behaviors, the HBQFA is relatively a short and easy questionnaire to use. The main limitation of this study was only applicable to female adolescents. Therefore, in future study including of male adolescents is important. In addition, since the study conducted among a sample of students from Tehran city, it cannot be generalized to people in other geographic locations or at other facilities. Test retest reliability assessment was limited to Tehran in Iran participants; hence further investigation of this property outside of Tehran is needed.
The results of this study showed that HBQFA questionnaire is valid and reliable instrument that can be used for research/survey, educational and practical purposes. On the other hand, because, the majority of subscales showed high internal consistency reliability, therefore, suggests good reliability instrument. Therefore, the findings of the current study suggest that theory-based HBQFA questionnaire is a valid and reliable instrument. The results of this study can be used in schools across the country.
This article is a part of a PhD thesis. The authors would like to acknowledge the contribution of Health Education and Promotion Office of Ministry of Health, Tehran University of Medical Sciences (TUMS) and all study area school authorities and managers. In addition, the author would like to appreciate the tolerance and cooperation of study participants.