preparing and analyzing data lesson chap 10

Module 8: Preparing and Analyzing Data

Chapter 10 - Data Analysis

On-line Lesson

The increased availability of statistical analysis software is making research projects easier to undertake.

Popular Statistical Packages include:

	Microsoft Excel
	SAS
	SPSS

Web Reference: Statistical Software List

Two Goals of Data Analysis

exploratory goal: to explore the data for any trends.
functional goal: finding answers to specific questions from the data.

Descriptive Analysis

Any form of data may provide descriptive information. The accepted "margin of error" for opinion and behavior surveys is .05%. This is also referred to as the "level of significance" (p value of .05).

Types of Descriptive Analysis

	Frequency, the total of times a score occurs in a distribution (item).
	Central tendencies, gives one score or measure that represents an entire group. Mean: the sum of the scores divided by the number of scores. *Median: the point that divides a distribution of scores exactly in half. Mode*: the most frequently occurring score (the score with the greatest frequency).
	Percentile measure
	Standard Deviation, the average distance of each score from the mean.

Analysis of Trends

review response frequencies to see how different groups answered a particular question using cross-tabulation.
comparing means of different groups.

Tests of Differences

Test of differences are used to determine if statistically significant differences exist between groups in the population.

Measurement Categories (each requires a different statistical analysis)

	measurement of the differences in count between two or more groups;
	measurement of the differences in mean attitude and opinion between two groups;
	measurement of the differences in mean attitude and opinion between three or more groups;

Statistical Analyses

	chi-square (x²)
	T-test (t)
	analysis of variance (ANOVA) one-way, computes the difference between groups.

Analysis of Relationships

Attempts to determine if any statistically significant differences indicate the existence of a causal relationship that can be statistically established.

correlation coefficient (rho)

Factor Analysis and Reliability Checks

Internal validity relates to whether a questionnaire is measuring what it is supposed to measure.

External validity refers to the generalizability to a larger population.

Cronbach's alpha

[Class] [Module 8] [Chapter 9 Lesson] [Chapter 11 Lesson]