
Joseph C. answered 03/28/19
SPSS Expert Explains Output & Helps APA Writeup & Dissertations
The answer to this question is yes, and yes;)
SPSS can be used for both descriptive and inferential statistics and has a robust set of features for both of these purposes.
Yes, Descriptive Statistic
SPSS has a robust set of "descriptive statistic" tools to describe the central tendency and dispersion of the data you are analyzing. There are also calculations in SPSS to examine the distribution of the data points in your dataset. This is particularly helpful in determining if your distribution is normally distributed, bi-modal, some other configuration.
- You can quickly and easily have SPSS calculate mean, median, mode and standard deviation, range.
- You can quickly and easily generate the frequency distribution of your data.
- You can test for normality with measures of skewness and kurtosis.
- You can visually represent your distribution or compare distributions with the histogram, box-plot, bar charts, and several charting options.
Yes, Inferential Statistics
SPSS includes a wide variety of inferential tests that can be used to answer questions of statistical significance that can be used for formal hypothesis testing. The most common tools for inferential statistics in SPSS will show differences in central tendency (t-test, ANOVA, MANOVA), and show how various variables/ attributes are associated (correlation, linear regression, logistic regression).
- You can compare 2 groups means with a t-test
- You can compare several groups means with 1-Way ANOVA, 2-Way ANOVA... or MANOVA
- You can visually demonstrate the differences between means using box-plots or bar charts.
- You can detect association of two variables with bivariate correlation.
- You can visually represent the association between two variables using scatter plots.
- You can build linear regression or logistic regression models to test how well one or more predictor variables (independent variable(s)) and if the prediction of the model is significantly better than chance.
- You can represent the line created by a linear regression in a scatter plot.