Person-Based Responsiveness Drives to Drop Resolution SPSS - Assignment Example

SPSS out-put data analysis report Student’s name Affiliated institution Part A Regression analysis and mixed-design ANOVA were conducted to evaluate disordered eating by the Australian Catholic University (ACU) . The investigated question matter is whether intuitive eating, a dependent variable has a superior effect on weight loss compared to restrictive eating. The effect of dichotomous thinking on intuitive eating after controlling for body image appreciation and healthy exercise is also investigated (Norusis, 2006). The data collected on a sample N=314 individuals are analyzed using SPSS. The mean and standard deviations of Intuitive eating: M=3.29,SD=0.86, body image appreciation: M=3. 10, SD=0.90 and exercising for health scale: M=5.38, SD=1.38 are small compared to those of Dichotomous thinking global score: M=25. 3, SD=8.36. This means that the observation made about these variables were related, i.e. those who ate when hungry most likely have a positive body image and were more likely to exercise for overall health benefits. The dichotomous thinking scale registered the highest mean and standard deviation. This implies that keeping the other variables constant, dichotomous thinking is inversely related to intuitive eating. Descriptive statistics for intuitive eating and other variables N=314 M SD N Intuitive eating 3.29 .86 314 Dichotomous Thinking global scale 25.3 8.36 314 Body appreciation Total score 3.10 .90 314 Exercise health scale 5.38 1.38 314 Correlation table The Pearson’s correlation was used to estimate the correlation among the variables. The correlation table summarizes the results. There is a statistically significant correlation, i.e. p> 0.05 between intuitive eating and all the variables (0.000).Intuitive eating and dichotomous thinking have a statistically significant correlation of p=0.00, meaning that with all the other factors controlled for, dichotomous thinking would affect intuitive eating negatively. There was an outlier in the relationship between dichotomous thinking global score and healthy exercise, p> 0.05 i.e. p=0.207. This was the only outlier and its effect on the data results is negligible. Outliers were diagnosed using the partial regression plot. The Pearson’s r is not discussed in the correlation context to avoid the possibility of false correlation given the data sample size (Contributors & Corbitt-Hall, n.d). To further examine whether the correlation observations made among the various variables under study were true, a scatter plot was run. The assumptions made when running the regression of the data set are: Linearity and additivity of the relationship between intuitive eating and other variables, statistical independence of errors and normality of the error distribution (Vacha-Haase et al., 2000). The regression coefficients showed significant p values of between (p=0.00, and p= 0.01 of intuitive eating and other variables. The ANOVA analysis showed a significant relationship between intuitive eating and the other variables, p= 0.00. The degrees of freedom of regression and residual are 1 and 312 respectively. Regression standardized residual against frequency histogram showed a normal distribution. Partial regression plots showed a positive correlation between intuitive eating and body image appreciation as well as with exercising for overall fitness. The partial regression plot showed a negative relationship between intuitive eating and dichotomous thinking. The results show that intuitive eating is a better way of controlling weight gain. The observations show that individuals who eat when hungry have a body image appreciation and exercise for overall health benefits. The effect of dichotomous thinking has a negative correlation with intuitive eating, keeping the other variables constant since it is directly proportional to restrictive eating hence disordered eating. The conclusion of the research is that intuitive eating is a better method of weight control as opposed to restrictive eating (Thomas et al., 2008). Since dichotomous thinking is used by the restrictive eaters which often leads to disordered eating in case the dieting rules are broken; the dichotomous thinking has an inverse relationship with intuitive eating. Part B Data analysis was conducted to investigate the hypothesis that person-based feedback leads to lower perseverance compared to process-based feedback. Independent variable of gender comprised boys n=45 and girls n=45 were subjected under 3 dependent variables of person-based, process-based and control conditions. In each of the conditions, boys: n=15 and girls: n=15.Under person_based category, boys: M=129, SD=22.3 performed better than girls: M=105, SD=31.6. This implies that boys could persevere longer than girls after a person- based motivation. The boys in this category also performed better overall compared to control: M=120, SD=36.4 and process_based group: M=123,SD=31.5. The girls: M=139,SD=29.6 performed better than boys: M=123,SD=31.5 under the process_based category. This postulates that process_based motivation positively impacted girls than boys. The girls in this group also performed better than the girls in the control and person_ based group: M=136,SD=30.3. In the control group, the girls could persevere longer than boys. The average for both the genders in the three groups show that, the process_based group persevered longer followed by the control group and finally the person_based group. This observation supports the expected findings of the study question, i.e. that the person_based feedback results in lower motivation and perseverance than process_based feedback. The table below is an ms excel file showing a bar graph with SE values. The data analyses run include the t-test which gives both the descriptive statistics and Leven’s test (Pallant, 2013). The null hypothesis tested is that the error variance of the dependent variable is the same across the groups (f). The p = 0.398 is more than 0.05 thus the null hypothesis is not rejected. The ANOVA to compare the dependent variable (condition and gender) and the independent variables was run, which yielded the test of between subjects results. The degrees of freedom in condition and gender and error are: df= 2,84 respectively. The results showed statistical significance in between some of the groups which prompted the conduction of post Hoc test (estimated marginal means). It is possible that the difference in perseverance among the sample was due to condition alone, gender alone or gender and condition together. There is an equal: SD= 7.892 across the groups with a (95% CI [89.639,154.875]). To further investigate the interaction between variables, a plot of condition against estimated marginal means; as well as gender against estimated marginal means were plotted. The Explore command was used to confirm and test the normality of the findings (Arkkelin, 2014). Although the dataset has less than 2000 elements which means the Shapiro-Wilk test would suffice, the Kolmogorov- Smirnov test was run as well (Kline, 2004). P=0.308, p=0. 200 are both higher than 0.05 implying no statistical significance. The skewiness of the data set were investigated using Q-Q plots and a box plot (Ghasemi& Zahediasl, 2012). The results support that process-based feedback has more motivation compared to person_based motivation as indicated by longer perseverance in the girl population alone. This hypothesis is rejected in the boy population where the person_based motivation yielded higher results compared to process_based motivation. The hypothesis holds true for combined means of both boys and girls. This means that the conditions, i.e. process based feedback and person based feedback are affected by other factors i.e. gender (Banerjee et al., 2009) References Arkkelin, D. (2014). Using SPSS to understand research and data analysis. Banerjee, A., Chitnis, U. B., Jadhav, S. L., Bhawalkar, J. S., & Chaudhury, S. (2009). Hypothesis testing, type I and type II errors. Industrial psychiatry journal, 18(2), 127. Contributors, G., & Corbitt-Hall, D. J. P-Values. Ghasemi, A., & Zahediasl, S. (2012). Normality tests for statistical analysis: a guide for non-statisticians. International journal of endocrinology and metabolism, 10(2), 486-489. Kline, R. B. (2004). Beyond significance testing: Reforming data analysis methods in behavioral research. Norusis, M. J. (2006). SPSS 15.0 guide to data analysis. Upper Saddle River, NJ: Prentice Hall. Pallant, J. (2013). SPSS survival manual. McGraw-Hill Education (UK). Thomas, R. P., Dougherty, M. R., Sprenger, A. M., & Harbison, J. (2008). Diagnostic hypothesis generation and human judgment. Psychological review, 115(1), 155. Vacha-Haase, T., Nilsson, J. E., Reetz, D. R., Lance, T. S., & Thompson, B. (2000). Reporting practices and APA editorial policies regarding statistical significance and effect size. Theory & Psychology, 10(3), 413-425. Read More