Exam 2 Flashcards by Jennifer Bigalke

what is a disadvantage and advantage of group design

Advantage- be more generalizable if equivalence is met for groups
Disadvantage- could potentially cost more and take more time

How well did you know this?

Not at all

Perfectly

what are the three treatment E’s?

Efficacy, Effectiveness, Efficiency

How well did you know this?

Not at all

Perfectly

the results of a study can be generalized to a people with similar characteristics

logical generality (external validity)

How well did you know this?

Not at all

Perfectly

What are the three factors that influence statistical outcomes?

effect size
variability of scores
sample size

How well did you know this?

Not at all

Perfectly

Data are approximately normally distributed is called what?

parametric statistics

How well did you know this?

Not at all

Perfectly

what are the four basic components of a methods section?

subjects/participants
experimental materials: DV(s), IV(s), testing materials
Procedure (step-by-step procedure to gather and analyze data)
data analyses/statistical analyses

How well did you know this?

Not at all

Perfectly

this also refers to generality: the results be
extended to other experimenters,
subjects, settings, etc

external validity

How well did you know this?

Not at all

Perfectly

non-parametric statistic method to determine correlation

contingency coefficient (C) & spearman rank (rho)

How well did you know this?

Not at all

Perfectly

What is the parametric statistic method to determine means in between-subject design?

independent t-test

How well did you know this?

Not at all

Perfectly

Measures the difference between two means (e.g., treatment vs. control group) relative to the pooled standard deviation

What effect size?

cohen’s d

How well did you know this?

Not at all

Perfectly

two IV;s, with two or more levels is what type of ANOVA?

two-way

How well did you know this?

Not at all

Perfectly

parametric statistic method to compare the means within design

paired t-test (two data sets) and repeated ANOVA/MANOVA (> 2 data sets)

How well did you know this?

Not at all

Perfectly

One threat to external validity is the Hawthorne effect. What is the Hawthorne effect?

The Hawthorne effect is when a participant may act differently when they know they are being “watched” in a study. It is an external validity threat to experiments.

How well did you know this?

Not at all

Perfectly

the results of a study can be generalized to the population

inferential generality (external validity)

How well did you know this?

Not at all

Perfectly

comparing the mean of two groups

independent t-test

How well did you know this?

Not at all

Perfectly

what is the effect of one of your IVs on the DV

main effect

How well did you know this?

Not at all

Perfectly

What are the key elements of the ‘discussion’ section in research articles (7 total)?

state major findings
explain meaning and important of the findings
related to findings of similar studies?
consider alternative explanations for findings
state the clinical relevance of the findings
study’s limitations
suggestions for further research

How well did you know this?

Not at all

Perfectly

how can you control external validity better?

subject randomization
larger sample size

How well did you know this?

Not at all

Perfectly

what is inferential statistics?

generalize from sample to population, make predictions

How well did you know this?

Not at all

Perfectly

compare the means of two groups

independent t-test

How well did you know this?

Not at all

Perfectly

taking a test once may affect subsequent administrations of that test. It is directly related to “Within or repeated measures”.

what internal validity?

reactive pretest

How well did you know this?

Not at all

Perfectly

more than two IV’s, with two or more levels is what type of ANOVA?

multi-way

How well did you know this?

Not at all

Perfectly

t-test, ANOVA, regression, CIs are examples of what type of statistics?

inferential

How well did you know this?

Not at all

Perfectly

what are the threats to internal validity?

History
Maturation
Reactive Pretest
Instrumentation
Statistical Regression
Differential subject selection
Attrition
interaction of Factors

How well did you know this?

Not at all

Perfectly

non-parametric statistic method to compare means between design

Mann-Whitney U test for two data sets

control group not matched to Tx group is what type of experimental design?

quasi-experimental

what should you include in the subjects/participants section of the methods section?

1. inclusion/exclusion criteria 2. sample size 3. IRB approval?

why is measuring baseline important for a study?

provides a point of reference to determine changes or improvements during the treatment. It helps ensure that any observed effects in the dependent variable are due to the manipulation of the independent variable(s) rather than other extraneous factors. Without a baseline, it is difficult to attribute changes solely to the treatment.

mean, median, mode, range, variance, SD are what type of statistics?

descriptive

what are the two experimental validities related to subjects?

internal and external

What are things to avoid in the discussion section?

1. overrepresentation of the results 2. unwarranted speculation 3. inflation of the importance of the findings 4. tangential issues

what are the threats to external validity?

1. subject selection bias 2. history 3. experimenter effect 4. Hawthorne effect 5. testing effect 6. situation effect 7. interactions between subject section bias and DV/IV

what are the three things efficacious treatments should have?

1. therapeutic effect 2. clinical significance 3. personal significance

What are the three major t-tests?

paired, independent, one sample

compare the mean of a sample to a predefined population mean (our class mean GPA vs mean GPA of all SLPs in the US.

one sample

parametric statistic method to compare the means between design

independent t-test and ANOVA/MANOVA

what is the difference from ANOVA design (in terms of the number of variables)?

ANOVA is for when there is one DV, MANOVA is for when there is more than one DV

compared the means of two levels only (i.e. pre and post in the single group

paired t-test

Regression towards the mean what internal validity?

statistical regression

what experimental validity makes sure you have unbiased and enough size of sample for generalization of your findings which is important in clinical study

external

A variation of Cohen's d that corrects for bias in small samples

hedeges' d

what is a disadvantage and advantage of single subject design

Advantage- can give a very detailed and in depth qualitative results (case study) Disadvantage- not generalizable

biological changes naturally occurring over time during the study What internal validity?

maturation

combination of other validity threats is what internal validity?

interaction of factors

when is a post hoc analysis necessary?

significant main effect and significant interaction effect

what section of a paper should provide the research design to answer the research questions (or hypotheses).

methods

if you predict a specific direction (e.g., an increase or decrease) for your hypothesis

one-tailed test

Students taking statistics courses in psychology at the Baylor University reported studying more hours for tests (M = 121, SD = 14.2) than did college students in Texas, t(33) = 2.10, p = .034. What t-test?

one sample t-test

non-parametric statistic method to compare means within design

Wilcoxon for two data sets

what are three types of effect sizes?

Cohen's d Hedges' g Pearson's r

compared the means of two levels only (pre and post in the single group)

paired t-test

what type of data do you use for non-parametric stats? -nominal -ordinal -interval -ratio

nominal and ordinal

what is the full name of the IRB?

institutional review board

matched control group is what type of experimental design?

true-experimental

What is advantage of using ANOVA over t-tests?

ANOVA controls for Type I errors if there are more than two groups, while t-tests do not. If you run a t-test multiple times there is a 5% increase of type I each time. ANOVA controls for this.

one IV with more than two levels is what type of ANOVA?

one-way

compare the mean of a sample to a predefined population mean

one sample t-test

When the distribution of your data deviates markedly from normal distribution or bell curve of normality

non-parametric statistics

what does p=0.05 mean?

5% chance that the change in the DV are not the result of the manipulation of the IV(s).

indicate the number of independent values that can vary in an analysis without breaking any constraints

degrees of freedom (df)

when subjects drop out of a study, that will affect interpretation of results. Recruit new subjects to be full sample or analyze data as they are without further recruitment what internal validity?

attrition

what is the full name of HIPAA

health insurance portability and accountability act

changes in measurement due to calibration issues, or changes in observers/scorers, etc what internal validity?

instrumentation

what is MANOVA?

multivariate ANOVA

parametric statistic method to determine correlation

Pearson (r) and Multiple regression analysis

he extent to which an experiment accurately measures what it intended to measure (i.e., relationship between IV(s) and DV)

internal

Measures the strength and direction of the linear relationship between two continuous variables

pearson's r

what does r2=.77 tell you?

77% of the relationship in the correlation is predicted for the IV and DV

what are the four most common statistical analyses in CSD?

1. t-test 2. ANOVA 3. correlation 4. regression **all parametric tests

no control group, or untreated controls is what type of experimental design?

pre-experimental

failure to balance extraneous factors across groups. It is directly related to “between subject design” what internal validity?

differential subject selection

When a main effect is found in data analysis, what analysis is required if there are multiple levels within the factor

post hoc

is a common measure of clinical significance used to estimate the differences between two groups (i.e., standardized mean difference) or the strength of an association between variables (e.g., odds ratio, relative risk ratio, Pearson’s coefficient)

effect size

this effect should be reported if you tested two or more IVs

interaction effect

extraneous events that occur between the first measurement and subsequent measurements What internal validity?

history

what type of data do you use for parametric stats? -nominal -ordinal -interval -ratio

interval or ratio