Lecture 15_independent samples t-test

Lecture 15_independent samples t-test - 10/25/2010...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
10/25/2010 1 Independent Samples t-test 1 Independent t-Test • The independent samples t-test requires no information about the population μ or σ – Evaluates differences in two sample means – Sample means on a continuous dependent variable are computed separately for each group and the two means are compared – Called a “between subjects design” because subjects participate in one group OR the other and you evaluate differences between the subjects 2 Example Research Questions • Do males and females differ with respect to statistics anxiety? • Do two curriculums result in different reading achievement for 6 th graders? • Is running or swimming more effective for weight loss? – All questions seek to answer whether observed mean differences between the group are greater than what you would expect to occur given chance alone 3
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
10/25/2010 2 Example • Wood et al. (1988) conducted a study that compared the effects of dieting and exercise on weight loss using a sample of 89 sedentary men. • Forty-two men were put on a diet, while the remaining 47 were placed on an exercise routine. – Dieting versus exercise weight loss regimens were compared 4 Framing the Research Question • If we were conducting this study, the research question could be written as follows: • “Are there weight loss differences in dieting versus exercise regimens?” – The word “differences” denotes a comparative question that demonstrates we are interested in comparing one group to another 5 Testing the Research Question • Previously, we compared a single sample mean to a known population mean (for both one sample z-tests and t-tests) • The independent t-test compares two sample means however, so we must alter the hypothesis to reflect this fact 6
Background image of page 2
10/25/2010 3 Previous Null Sampling Distributions • For the one sample tests the null sampling distribution was centered at the population mean of the variable – Scores in the distribution were sample means of different samples 7 Previous Null Sampling Distributions • For the dependent samples tests the null distribution was centered around 0, the population mean difference across pre- and post-test measurements – Scores in the distribution were sample mean differences of different samples D 8 New Null Sampling Distribution • The null sampling distribution for the independent t-test is also centered around a mean difference of zero, but now scores in the sampling distribution are the difference in two means across different samples 1 - 2 = 0 Sampling distribution contains mean differences across two groups 9
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
10/25/2010 4 Graphical Representation • If the null hypothesis is true, we expect the difference between the two different sample means will be zero • Large differences between the group means would be unlikely to occur in the null distribution 1 - 2 = 0 0 10 Null Hypothesis for the independent t-test • The null hypothesis is that the population means for the two groups are equal:
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/17/2011 for the course PYSC 227 taught by Professor Fairchild during the Spring '10 term at South Carolina.

Page1 / 14

Lecture 15_independent samples t-test - 10/25/2010...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online