Chapter 7: Single-Factor Designs Flashcards
a single factor design has..
one independent variable and can have 2 or more levels
What are the 4 types of single factor designs?
- independent groups
- matched groups
- repeated measures
- ex post facto
What are the questions one must consider when choosing a single-factor design?
- is the IV between- or within-subjects?
- is the IV manipulated or a subject variable?
- if IV is manipulated & BS, will you form equivalent groups by matching or by random assignment?
- if IV is WS, how many times is it tested per condition?
For which single-factor design woud you form equivalent groups using random assignment?
independent groups single-factor design
What do we know about a study if it uses an ex post facto single factor design?
The IV is between-subjects and is a subject variable
What designs can we use if we have a between-subjects manipulated IV?
matched groups or independent groups designs
What design is appropriate for a within-subjects IV?
a repeated measures design
What are the assumptions for a t-test?
- Approximate normal distribution
- Homogeneity of variance: similar variance of each set of scores being compared
when performing a t-test, you can chose between an ____ and a ____ t-test
independent samples, paired (dependent) samples
when is an independent samples t-test appropriate?
When the groups being compared are completely independent of each other (i.e., independent groups design and ex post facto design)
when is a paired samples t-test appropriate?
if the IV is a within-subjects factor (i.e., in a matched groups design or a repeated-measures design)
sWhat is an advantage of single-factor multi-level designs?
they can discover nonlinear effects
what becomes possible with within-subjects multilevel designs?
you can counterbalance single-factor, repeated-measures designs in more than one ways
When is it more effective to use a graph?
if there are large differences, if nonlinear effects occur, or if the result is an interaction between two factors
When is it preferred to use a table?
when data points are so numerous that a graph would be uninterpretable or when the researcher wishes to inform the reader of precise values
What is a risk that comes with performng multiple t-tests?
performing a type I error
a line graph is appropriate for presenting…
continuous variables
a bar graph is appropriate for presenting…
discrete variables
A type I error occurs when…
you reject the null hypothesis when it’s true
a type II error occurs when…
you accept a null hypothesis that isn’t true
What does a one-way ANOVA look for?
an overall significant effect that could exist somewhere between the levels of the independent variable
What is a post-hoc analysis used for?
to determine which level of an IV produced the significant effect that was found in the ANOVA
What influences the selection of post hoc analysis method?
sample size
how conservative the researcher wishes to be when testing for differences
What inferential statistic does a one-way ANOVA produce?
An F-score
for which designs can we use a one-way ANOVA for independent groups?
multilevel independent groups design
multilevel ex post facto design
for which deisgns can we use a one-way ANOVA for repeated measures?
multilevel matched groups design
multilevel repeated-measures design
What is a yoked control group?
when subjects in the experimental group participate for varying amounts of time or are subjected to different types of events in the study, each member of the control group is matched (yoked) to a member of the experimental group so that on average the time spent participating or the types of events encountered are kept constant