**IST:622 Term 4**

## Week 3 Objectives

- Identify different samples and determining which statistical procedure to use for data analysis
- Interpret meaning of scores, measurement and scientific inquiry
- Develop an appropriate evaluation plan for a given project
- Videos-
- Independent sample t-test
- MAC t-test 2 sample for means unequal
- MAC t-test sample paired difference

- Reading Assignments
- Russ-Eft & Preskill: Chap 5
- Pedhazur & Schmelkin Article
- Paired Sample t-test example
- Independent sample t-test example
- Lecture- Inferential Statistics (Week 3) (See accordion below)

**Assignment**

Follow instructions from Data File and Lab Instruction

Practice Questions-

- SAT scores of 100 girls compared to SAT scores of 100 boys in a private school in California. Are these dependent or independent samples? What is the degree of freedom? (Answer: These are independent samples because there is no indication of matching, pairing, or repeated measure. Therefore, the df=100+100-2=198)
- Twenty-five first graders' mid-term and end-of-term math exam scores are compared. Dependent or independent sample? What is the degree of freedom? (Answer: These are dependent samples because each student's mid-term score is compared to his/her end-of-term score. This is an example of repeated measure. Therefore, the df=25-1=24)

**Lecture Notes**

Inferential Statistics (Week 3)

- Simple random sampling
- Each member of the population has an equal and independent chance of being chosen
- The sample should be very representative of the population
- Systematic sampling
- The term nth stands for a number between 0 and the size of the sample that you want to select
- Stratified Sampling
- Select a sample that is representative of the population
- Use this method when the individuals in the population are not equal to...
- Convenience Sampling
- Captive or easily sampled population
- Not random
- Weak representativeness
- value of the means of each sample
- value of the standard deviations
- number of cases in each group
Independent vs. dependent sample- - Example of independent samples: randomly choose from two populations, use random selection procedure to assign elements from one population to two samples, cases in the two groups are not related or matched, etc. Basically, obtained scores in one group are independent or not related to the scores in the other group.
- Example of dependent samples: repeated measures using the same subjects, subject matching, selecting pairs of business partners, etc. Comparing pre-test and post-test scores after an instructional intervention to a group of students would need dependent sample t-test since it meets the definition of repeated measure
How Inference Works t-test for dependent samples- Is a process for determining if there is a statistically significant difference between the means of two dependent samples. (slide 19) For additional information review examples at: http://www.uwsp.edu/PSYCH/stat/11/hyptest2s.htm Summary- Inferential statistics tell us how much confidence we can have when we generalize from a sample to a population. In choosing the proper inferential statistic, an analyst must consider the size and number of samples drawn and the nature of the measures taken. |