SPSS Results Dissertation Results Examples
Statistical test used and Justification:
Data collected from the stroop test was analysed using a regression model. The main aim of the model was to determine whether individuals who show more interference in stroop test also show slower object naming times. For the regression model, the dependent variable used is object-naming time (overallobrt). In order to determine the independent variables that will be put into the model it is important to carry out a correlation analysis of the variables in the study. Once the correlation model has been carried out, any variable with a correlation coefficient of above 0.7 or 0.8 will be included in the model. The reason this model was selected is because it will provide an insight into the relationship between the dependent and independent variables.
Observed values and significance level
Secondly, examining table 2, it is also evident that stroopcongruent and object-naming time (overallobrt) have a weak correlation with a correlation coefficient of 0.372. This relationship is also evident in the scatter plot (chart 2). However, it will be included in the regression model since it shows a moderate to strong correlation to stroopincon with a correlation coefficient of 0.647. Furthermore, based on the p (significance value) of 0.003 from table 2 below it is clear that stroopcongruent is significant when it comes to predicting stroopincon since the p value is less than alpha (0.05). Therefore, it will be included in the regression model.
Examining the ANOVA output (table 3 above) the results show that the independent variables selected for the regression analysis are significant at the 5% level of significance since the p (significance value) is 0.043. The reason for this observation is that the significance value is less than alpha (0.05). This significance value indicates that the regression model coefficients are greater than zero. The variable stroopnut (color naming) will not be included in the model since it has the lowest correlation coefficient.
Examining the regression output table (table 5 above) it is possible to develop a regression model based on the data collected and analysed. The regression model can be predicted as follows:
Overallobrt (predicted) = 1.5 (stroopincon) – 0.097 (stroopcongruent) - 216.375
The model shown above can be used to predict object-naming time (overallobrt) based on stroopincon and stroopcongruent figures.
In conclusion, based on the coefficients above the following can be concluded:
A unit increase in stroopincon will lead to a 1.5 increase in overallobrt
A unit increase in stroopcongruent will lead to a 0.097 decrease in overallobrt
Chart 1: Scatter plot showing the relationship between Overallobrt and stroopcongruent
Chart 2: Scatter plot showing the relationship between Overallobrt and stroopincon