Showing 7 results for Factor Analysis
Ma Pourhosseingholi, Y Mehrabi, H Alavi-Majd, P Yavari,
Volume 1, Issue 3 (2-2006)
Abstract
Background and Objectives: Logistic regression is one of the most widely used generalized linear models for analysis of the relationships between one or more explanatory variables and a categorical response. Strong correlations among explanatory variables (multicollinearity) reduce the efficiency of model to a considerable degree. In this study we used latent variables to reduce the effects of multicollinearity in the analysis of a case-control study.
Methods: Our data came from a case-control study in which 300 women with breast cancer were compared to 300 controls. Five highly correlated quantitative variables were selected to assess the effect of multicollinearity. First, an ordinary logistic regression model was fitted to the data. Then, to remove the effect of multicollinearity, two latent variables were generated using factor analysis and principal components analysis methods. Parameters of logistic regression were estimated using these latent as explanatory variables. We used the estimated standard errors of the parameters to compare the efficiency of models.
Results: The logistic regression based on five primary variables produced unusual odds ratio estimates for age at first pregnancy (OR=67960, 95%CI: 10184-453503) and for total length of breast feeding (OR=0). On the other hand, the parameters estimated for logistic regression on latent variables generated by both factor analysis and principal components analysis were statistically significant (P<0.003). The standard errors were smaller than with ordinary logistic regression on original variables. The factors and components generated by the two methods explained at least 85% of the total variance.
Conclusions: This research showed that the standard errors of the estimated parameters in logistic regression based on latent variables were considerably smaller than that of model for original variables. Therefore models including latent variables could be more efficient when there is multicollinearity among the risk factors for breast cancer.
M Yazdi, H Mahjub,
Volume 7, Issue 1 (6-2011)
Abstract
Background & Objectives: Maternal health status is one the important health index since it could be a hallmark of
performance of health care service particularly in rural area. The aim of this study is ranking and identifying
homogeneous provinces based on maternal health indices in Iran.
Methods: Based on the indices in the vital horoscope which were reported by Iran Ministry of Health in 2006,
fifteen indices related to maternal health have been selected. Using factor analysis, a factor that can represent
rural maternal health in provinces was chosen and provinces ranked according to it. A hierarchical cluster analysis
employed for identifying homogenous province based on a distance matrix obtained from the same selected
indices in factor analysis.
Results: Factor analysis demonstrated that 89% of data variation included in this model. Based on an
appropriate factor as index of maternal health, ranking was run in all provinces in Iran. According to this model,
the maternal health status was in good condition in the rural areas of Gilan, Mazandaran and Tehran provinces,
respectively. While maternal health were in inappropriate condition in rural areas of Khuzestan, Kohkiluyeh and
Hormozgan provinces. The worst rural area was the province of Sistan and Baluchestan.
Conclusion: It could be concluded that employing multivariate analysis would be an effective tool for ranking
and classifying homogeneous population (provinces) in terms of maternal health indices. Therefore health policy
makers should consider the practical method of this analysis for any decision making.
Sj Tabibi, Z Rezaie Mohammadi , T Allahviranloo, Mh Taghdisi,
Volume 10, Issue 3 (12-2014)
Abstract
Background & Objectives : Leadership power and management are discussed as sensitive and key elements in the safety issue. A safety leader can provide the necessary incentives for hard work, efficiency, and more productivity in the organization which includes various working groups and also strengthening a sense of individual responsibility to promote the safety performance in order to achieve the organizational goals. This study aimed to determine the validity and reliability of the Safety Leadership Questionnaire in the vehicle manufacturing industry.
Methods : This cross-sectional study was conducted on 260 employees of Pars-Khodro Company in the manufacturing lines. All samples were selected randomly. The validity of the questionnaire was determined by statistical methods such as exploratory factor analysis, confirmatory factor analysis, and the heterogeneity between questions. Also, the Cronbach's alpha was used in order to investigate the reliability of the questionnaire.
Results: The results of exploratory factor analysis showed three factors including decision making, control, and incentives explained 66 percent of the total variance. Confirmatory factor analysis represented the appropriate fitness of information with a three-component structure (P<0.001, df = 2, 𝜒 2 = 5.36, CFI = 0.957, RMSEA = 0.089). The Cronbach's alpha coefficient was higher than 0.7 for all three factors.
Conclusion : In this research, decision making was an important factor in measuring safety leadership. Moreover, the present study represented an indicator which could be used in the other surveys of safety and industrial psychology.
Ma Soleimani, S Zarabadi Pour , Y Yaghoobzadeh, S Pahlevan Sharif , H Sharif Nia ,
Volume 14, Issue 1 (6-2018)
Abstract
Background and Objectives: The effect of heart diseases on the quality of life is the issue needs attention of health care providers. Improving quality of life is considered as the goals of rehabilitative therapies. This study conducted to evaluate the McGill Quality of Life Questionnaire in patients with heart diseases.
Methods: In this cross-sectional study, 500 patients with heart diseases were recruited from BooAli Sina Hospital and Velayat Hospital affiliated with Qazvin University of Medical Sciences from May to August 2016. The participants completed the McGill Quality of Life Questionnaire. The construct validity (including convergent and discriminant validity) and Reliability using the Cronbach’s alpha, theta, and McDonald's Omega of the McGill Quality of Life Questionnaire were evaluated. The structure of the Questionnaire was assessed using factor analysis.
Results: Three factors, including overall view of the quality of life, physical aspect, and psychological dimension, were extracted. Model fit indexes confirmed a good fit of he McGill Quality of Life Questionnaire (Comparative of Fit Index: CFI=.918, incremental fit index: IFI=.919, Adjusted Goodness of Fit Index: AGFI=.844, RMSEA=.079, Minimum Discrepancy Function by Degrees of Freedom divided: CMIN/DF=2.97, Parsimonious Normed Fit Index: PNFI=.681, Parsimonious Comparative Fit Index: PCFI=.709). Convergent and divergent validity, internal consistency, and construct reliability of the questionnaire were confirmed.
Conclusion: The findings revealed that the three-factor model of the McGill Quality of Life Questionnaire has satisfactory validity and reliability. Thus, this questionnaire can be used in future studies to assess the quality of life of patients with heart diseases.
Mm Vakili, A Moazen, S Sharifi, K Kamali,
Volume 14, Issue 3 (12-2018)
Abstract
Background and Objectives: The results of recent studies suggest that model-based health education programs are more effective in changing health behaviors. Interventional programs based on health education theories can be used to explain individual and psychological behavioral determinants in order to reduce the risk of HIV/AIDS. This study was conducted with the aim of psychometric analysis of an HIV scale based on model, population and culture of Iran.
Methods: We used qualitative and quantitative methods during August 2015 to December 2017. The items were generated based on a literature review and interviews with an expert panel. After providing a primary list of the instrument items, we used content and face validity. The first step of the study was done in a sample of medical and non-medical college and high school student from Zanjan. The exploratory factor analysis, internal consistency reliability, and test-retest reliability of the scale were measured.
Results: Fifty-eight items were generated from interviews with experts and review of the literature. After measuring the face and content validity, the items decreased to 53 items. The exploratory factor analysis discovered five factors (perceived susceptibility, severity, benefits, barriers, and self-efficacy), including 30 items for high school students and 31 for university students, that predictive more than 58% of the observed variance. The Cronbach’s alpha coefficient for the sub-scales ranged from 0.80 to 0.86, and the ICC was above 0.86.
Conclusion: The results showed that the HBM was a valid and reliable tool for measuring the HIV/AIDS beliefs and could be used in future investigations.
M Eftekhar Ardbili , M Taban, M Hatamizadeh, H Ranjbar,
Volume 15, Issue 2 (9-2019)
Abstract
Background and Objectives: The Problematic Internet Use Questionnaire (PIUQ) is a scale for evaluating the adverse effects of high Internet usage. It was adopted into Persian in a previous study. The factorial structure of the adopted version was different from the original English version. The purpose of this study was to determine the factorial structure of the Persian version of the Problematic Internet Use Questionnaire.
Methods: In this methodologic study, sampling was carried out in students of Iran University of Medical Sciences, University of Tehran, and Sharif University of Science and Technology. A total of 428 students completed the Problematic Internet Use Questionnaire. Exploratory factor analysis was performed using SPSS. Confirmatory factor analysis was carried out using AMOS. The reliability of the Scale was determined using Cronbach’s alpha coefficient (α=0.89).
Results: Exploratory factor analysis showed four factors including online preoccupation, adverse effects, social interactions, and excessive use. The mean ± SD of the domains was 2.12±0.81, 2.01±0.86, 2.18±0.85 and 2.39±0.97 for online preoccupation, adverse effects, social interactions, and excessive use, respectively.
Conclusion: The exploratory and confirmatory factor analysis showed that the Persian version of the Problematic Internet Use Questionnaire had a relatively similar structure with the English version. One dimension of the questionnaire was split to two new dimensions. This questionnaire can be a useful tool for measuring problematic behaviors related to the excessive use of the Internet.
Fereshteh Eidy, Hoosein Fallahzadeh, Rahman Panahi, Jamshid Jamali,
Volume 18, Issue 3 (12-2022)
Abstract
Background and Objectives: Today, overweight and obesity are among the significant challenges in the world. They can have adverse effects on quality of life. Quality of life is determined by social and physical environment. The present study investigated the construct validity of the Persian version of the Impact of Weight on Quality of Life (IWQOL-lite) questionnaire using multilevel confirmatory factor analysis.
Methods: The present study was conducted on 310 people over 18 years old living in Yazd city. The data were analyzed using m-plus6.2 software.
Results: The mean age of the participants in this study was 33.97± 4.70 years. Sixty percent (N=183) of subjects were male. The indices of the one-level model were more appropriate than those of the two-level model, and the one-level confirmatory factor analysis model had a good fitting to the data (CLI: 0.98, TLI: 0.98, RMSEA: 0.038).
Conclusion: The results of this study indicated the need for more effectiveness of urban areas on quality of life. The one-level confirmatory factor analysis model confirmed the construct validity of the IWQOL-lite questionnaire. This questionnaire can be used in the Iranian population.
