Lab 8 Inferential Statistical Analyses (Part C) – SPSS RELIABILITY ANALYSIS OF THE SCALE Introduction Reliability is a measure of the internal consistency of a set scale items. It takes the value between 0 and 1. If the value come close to 1 (e.g. 0.75, 0.85), the more reliable a set of scale we found. There are a number of different coefficients. Most commonly used is Cronbach’s alpha, which is based on on the average correlation of items within a test if the items are standardised. Cronbach’s alpha can also be interpreted as a correlation coefficient. Cronbach alpha values quite sensitive to the number of items in the scale. With short scales (e.g. scales with fewer than ten items), it is common to find quite low Cronbach values. Procedure for checking the reliability: Click on Analyze Scale Reliability Analyis Click on the all individual items that make up the scale, and move these in the box marked Items. In the Model section, make sure Alpha is selected. Click on the Statistics button. In the Descriptives for section, click on Item, Scale, and Scale if item deleted. Click on Continue, then OK. Example of output: 51
How to interpret: Check that the number of items is correct. The most important figure is Alpha value. This is Cronbach’s alpha coefficient, which in this example, the value is 0.89. This value is above 0.70, so the scale can be considered reliable with the sample. Column marked Corrected Item-Total Correlation give you an indication of the degree to which each item correlates with the total score. Low values (less than 0.30) in this example indicate that the item is measuring something different from the scale as a whole. If the scale’s overall Cronbach alpha is too low (e.g. less than 0.70), you may need to consider removing items with low item-total correlations. Column Alpha if Item Deleted the impact of removing each item from the scale is given and you can compare these values with the final alpha value obtained. If any values in this column are higher than final alpha value, you may want to consider removing this item from scale. VALIDITY ANALYSIS OF THE SCALE CORRELATION Inferential Statistics Exploring relationships among variables / Association among factors Correlation, regression Pearson Correlation Correlation analysis is used to describe the strength and direction of the linear relationship between variables. Pearson correlation used when you want to explore the strength of the relationship between two continuous variables. It also can be used when one the variables is dichotomous; that is only have two values (e.g. sex). It gives an indication of both direction (positive or negative) and the strength of the relationship. Pearson correlation coefficients (r) can take on only values from -1 to +1. What do you need both continuous variables or one continuous variable and the other one dichotomous variable Procedure: Analyze Correlate Bivariate Select the two variables and move them into the box marked Variables Check the Pearson box and the two-tailed box 52
Click on Options you can also obtain mean and SD if you wish. Click Continue and OK. Example of output: Spearman Rank Order Correlation (rho) Non parametric alternative and used to calculate the strength of the relationship between two continuous variables. The same basic procedure is used to request both Spearman correlation (rho) and Pearson’s r. The only difference is the option that is ticked in the section labelled correlation coefficients. 53
Search
