PSYC 2260 Lecture Notes - Lecture 7: Effect Size, Psychological Intervention, Random Assignment
26 Jun 2018
School
Department
Course
Professor
PSYC 2260 Introduction to Research Methods in Psychology Chapter 7
Chapter 7 Introduction to T-tests: Single Sample and dependent Means
The T-test for a Single Sample
In reality, researcher work with sample information, population variance it’s unknown
- Thus: have to estimate it from the sample data
Biased estimate: range of sample is smaller than the range in population. So based on sample
data is an underestimate of population SD
To correct the bias adjust the formula:
- the dominator, N-1, is called the “degree of freedom” (df)
- E.g. total = 10units of X from 3 people
- df = 3-1 = 2
Important note: alternative formulas
- Sample description
- Population estimate = S2 = (note that: SS = square of sum; df = N-1)
Using N-1 has the effect of slightly increasing the variance
- Un-biases it and make it more like the population variance
- Note: SPSS uses “N-1” in computing SD2 and SD
Variance of distribution of means:
Standard deviation of the distribution of means:
Instead of calculating Z, we calculate ‘t” (use SM instead of σM)
T distribution is not normal curve
- Heavier tails, more scores on extreme
- The more degree of freedom, the shape is more closer to the normal curve
- The cut off scores are more extreme
T-test for Dependent Mean
“repeated measures/ within-subjects”: Used to compare 2 scores coming from the same people
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Chapter 7 introduction to t-tests: single sample and dependent means. In reality, researcher work with sample information, population variance it"s unknown. Thus: have to estimate it from the sample data. Biased estimate: range of sample is smaller than the range in population. So based on sample data is an underestimate of population sd. To correct the bias adjust the formula: the dominator, n-1, is called the degree of freedom (df) E. g. total = 10units of x from 3 people df = 3-1 = 2. Population estimate = s2 = (note that: ss = square of sum; df = n-1) Using n-1 has the effect of slightly increasing the variance. Un-biases it and make it more like the population variance. Note: spss uses n-1 in computing sd2 and sd. Instead of calculating z, we calculate t (use s m instead of m) The more degree of freedom, the shape is more closer to the normal curve.
