Research methods

1
Research methods Longitudinal data Statistics

• Patrick Harris, M.S.
• Research Planning
• Wyoming Department of Workforce Services
• patrick.harris1_at_wyo.gov
• 307-473-3811

2
Inferential Statistics

• One-Way Analysis of Variance (ANOVA)
• Used to examine differences in the dependent
  variable
• Population mean difference test
• Between-group variance and within-group variance
• Produces an F statistic

3
Inferential Statistics

• ANOVA

Difference
Population 2
Population 1
300
250
4
Inferential Statistics

• Multivariate Analysis of Variance (MANOVA)
• Similar to ANOVA except analyzes more than one DV
• DVs must be theorized to be related to one
  another

5
Binary Logistic Regression

• Similar to multiple regression except the
  dependent variable is binary (dichotomous) in
  nature
• d.v. has only two levels (e.g., pass/fail, drop
  out/no drop out)
• Independent variables are sometimes referred to
  as covariates
• Calculates the probability of group membership

6
Propensity Score Matching

• Involves the use of binary logistic regression to
  match individuals who received treatment with
  individuals who did not (the control group)
• A statistical procedure designed to pair
  individuals from each group on similar
  characteristics (LaLonde, 1986)
• Used to control for background characteristics in
  the absence of randomization

LaLonde, R. (1986). Evaluating the econometric
evaluations of training programs with
experimental data. American Economic Review, 76,
604-620.
7
Cluster Analysis
Fiske, S., Cuddy, A., Glick, P., Xu, J. (2002).
A model of (often mixed) stereotype content
Competence and warmth respectively follow from
perceived status and competition. Journal of
Personality and Social Psychology, 82, 878-902.
8
Factor Analysis

• Exploratory Factor Analysis (EFA)
• Technique used to analyze the underlying pattern
  of correlations among a set of measured
  variables.
• It is exploratory because the research does not
  hypothesize how the variables will correlate
• Extracts a smaller set of variables called
  factors
• Linear combinations are achieved through rotation
  in 3D space
• Self-Esteem Scale factor matrix

9
Factor Analysis
Rotated Component Matrixa Rotated Component Matrixa Rotated Component Matrixa Rotated Component Matrixa Rotated Component Matrixa Rotated Component Matrixa
  Component Component Component Component Component
  1 2 3 4 5
Reversed inferior to others at moment .756 .159 .183 .078 -.192
confident about my abilities .744 .183 -.091 .201 .113
Reversed frustered about performance .713 .121 .343 -.191 .150
Reversed trouble understanding reading .678 -.090 .028 .039 .376
Reversed i'm not doing well .585 -.013 .293 .247 -.090
Reversed less scholastic ability right now .575 .171 .105 .472 -.066
i freel confident i understand things .569 .034 .044 .556 .147
satified with my body looks right now .022 .829 -.074 .072 .107
i am pleased with my appearance .229 .813 .083 .132 -.179
Reversed dissatisfied with weight -.098 .735 .092 -.016 .208
Reversed feel unattractive .143 .613 .299 .262 -.328
i feel good about myself .313 .572 .213 .315 .078
Reversed displeased with myself .429 .469 .439 .113 -.085
Reversed concerned about impression .043 -.084 .756 -.189 .266
Reversed worried about looking foolish .212 .021 .696 .365 .011
Reversed worried what other people think .179 .350 .626 .099 .027
Reversed self-conscious .069 .309 .558 .168 -.426
feel as smart as others .355 .126 -.051 .709 -.057
feel that others respect and admire me -.154 .281 .274 .706 .215
Reversed worried success or failure .211 .207 .221 .234 .679
10
Factor Analysis

• Confirmatory Factor Analysis
• Similar to exploratory except the research is
  confirming the hypothesized factors
• CFA is theory-driven

11
End of Section Four

• Next Section
• Longitudinal Data Systems

12
Research Planning Wyoming Department of
Workforce ServicesP.O. Box 2760246 S. Center
St. Casper, WY 82601(307) 473-3807 DWS-RESEARCH
PLANNING_at_wyo.gov http//wyomingLMI.org