Analyze Quantitative Data Easily
See important notes below!
Do not rush to conclusions until all results are in!
 One must organize and find meaning in the mass of collected data.
 Data must be looked at holistically
 Quantitative data are summarized statistically or mathematically
For statistical analyses, the two common types are descriptive and inferential.
Descriptive (used to describe and summarize data) Four descriptive statistics that are most useful to action research: 1. Mean (average): How a typical person scored on the test/survey 2. Standard Deviation: Show much mush scores vary from each other.  High SD: significant range in results  Small SD: similar range of result 3. Percentage: Shows the number of individuals who gave a particular response to achieve a particular score  *Computers are your friend!
 Inferential: This tells us how much confidence we can have in generalizing from a sample to a population.

Descriptive (used to describe and summarize data)
1. Mean (average): How a typical person scored on the test/survey
2. Standard Deviation: Show much mush scores vary from each other.
 High SD: significant range in results
 Small SD: similar range of result
3. Percentage: Shows the number of individuals who gave a particular response to achieve a particular score
4. Correlation coefficient(r): A measure of relationship indicating the degree to which 2 or more variables are related.
*Computers are your friend!
 Easier to use, faster, easily accessible!
 No need to remember formulas or equations!
 STATPAK: userfriendly and FREE!
 Download: http://wps.prenhall.com/chet_airasian_edresearch_8/38/9865/2525601.cw/content/index.html
Inferential:
 TTest: determines the significance of a difference between the means of two groups.
 Sign Test: Determines the significance of a difference between the means of one group.
 Mann Whitney UTest: determines the significance if two small groups differ significantly.
 ChiSquare Analysis: Determines the relationship between two or more nominal variables.
See below for types of inferential statistics:
TTest
 Used in groups larger than 15 people
 Determines the difference between 2 means of statistical significance
 Input raw scores into STATPAK
 A high level of significance shows the results can be attributed to the treatment and not chance error.
Sign Test
 Determines if post test scores are different from pretest scores.
 Example: Comparing effects of 2 different reading programs in one class.
 Used on one group designs.
 Must use a "sign table" to get probability value
 *Weakness: No control group.
MannWhitney UTest
 Can show if there is a significant difference between post test scores of 2 groups
 Involves control and experimental group
 Similar to TTest
 Uses pretest and post test scores
 Must consult a "U Table" to receive results
ChiSquare
 Used for nominal data (gender, ethnicity, socioeconomic status)
 Example: Relationship between gender and teacher morale?
 Can be calculated in STATPAK
Spearman Rho Shows the degree of the relationship between 2 sets of scores collected from the same group. * Useful to teachers because it can be used with small groups of subjects  Pearson r
 Analysis of variance ANOVA or FTest
**Also, ANCOVA (analysis of covariance)  used when similarities between groups cannot be determined at the beginning of the project.  adjusts pretest scores, so groups can be treated similarly 
Spearman Rho
* Useful to teachers because it can be used with small groups of subjects
Pearson r
 Used to determine relationship between 2 variables
 **Positive correlation: one variable increases, other increases
 **Negative correlation: one variable increases, other decreases