University of Tabriz Journal of Modern Psychological Researches 2322-3340 10 38 2015 07 23 Prediction of Marital Satisfaction Level of Female Students based on Attachment Styles: Comparing the Power of Logistic Regression and Artificial Neural Networks Prediction of Marital Satisfaction Level of Female Students based on Attachment Styles: Comparing the Power of Logistic Regression and Artificial Neural Networks 99 117 4082 FA Hossein Purshahriar Journal Article 2015 12 13 The aim of the present study was to compare the power of logistic regression and artificial neural network in the prediction of marital satisfaction of female students based on their attachment styles. Data were collected through the marital satisfaction questionnaire (ENRICH) and adult attachment Inventory that was completed by 300 female married students and were analyzed by the mentioned methods. Proportions of correct prediction of the two methods were compared using McNemar test as well. Results showed that although the artificial neural network model was more successful than the other, reducing number of the predictors into three items, there was no significant difference between the mentioned models. Hence, it could be proposed that when the number of predicting factors is in creased and goes beyond some levels, the artifical neural network model would be more successful than the logistic regression model. The aim of the present study was to compare the power of logistic regression and artificial neural network in the prediction of marital satisfaction of female students based on their attachment styles. Data were collected through the marital satisfaction questionnaire (ENRICH) and adult attachment Inventory that was completed by 300 female married students and were analyzed by the mentioned methods. Proportions of correct prediction of the two methods were compared using McNemar test as well. Results showed that although the artificial neural network model was more successful than the other, reducing number of the predictors into three items, there was no significant difference between the mentioned models. Hence, it could be proposed that when the number of predicting factors is in creased and goes beyond some levels, the artifical neural network model would be more successful than the logistic regression model. https://psychologyj.tabrizu.ac.ir/article_4082_6a464b143103a90771b3675ad6d64ccf.pdf