Child Development 133 
Research Methods in Human Development

Sections 02 & 06
Hembree            Fall, 2009

 

  Analyzing data

 I. Introduction

A. Conceptual perspective

1.      Comparing the means of the different groups

 

 

2.      Mean differences can be due to IV, confound, or error

 

 

3.      Comparing the results (means) we obtain with what we might expect by chance

 

B. The role of probability

1.      Chance and error

 

 

2.      Could these results have occurred by chance?

e.g.,  consider two means 12.0 and 13.0

 

 

 

All things being equal, we are more convinced when we see larger differences between means.  Why?

 

II. Hypothesis testing

A. Null hypothesis/Experimental hypothesis

 

 

 

B.  Fail to reject vs. reject

 

 

 

  • statistical significance

 

 

C.     Making decisions and making mistakes

Possible outcomes:

 

 

  Decision

 

 

Reject Null

Fail to reject Null

 

 

 

What is true in

the population?

 

Null is false

 

CORRECT

(Power)

 

 

Type II error

(β)

 

Null is true

 

Type I error

(α)

 

 

CORRECT

 

1.      Type I error 

 

 

2.      Type II error

 

 

 

Which error is worse?

 

 

 

3.      Power

 

 

 

Power is increased by:

  •  Reducing error

 

  •  Increasing # of subjects

 

  •  Increasing the effect size

 

 

 

 

III. Inferential Statistics

A. What do we mean by inference?

 

B.  t-test

t = association/lack of association

1.      calculate means

2.      calculate variance

3.      compute t

4.      find critical t, given specified alpha

5.      compare obtained t with critical t; make decision

 

 

 

degrees of freedom

 



 

C. Chi-square (handout)

  •  Used to test the association between two nominal (categorical) variables

 

 

 

 

 

 

Send problems, comments or suggestions to: hembrees@csus.edu

California State University, Sacramento
College of Education
Department of Child Development

Updated: August, 2009