Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

stats > Stats Final > Flashcards

Stats Final Flashcards

(54 cards)

Start Studying
1
Q

Describe the nature and purposes of descriptive versus inferential statistics.

A

Descriptive: describe/summarize data
Inferential: test hypothesis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Describe what the mean assesses

A

The average of scores

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Describe what the standard deviation assesses

A

The average deviation from the mean

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Describe the features of the z-score distribution. Describe the metric problem and how z-scores help overcome this problem.

A

The mean is 0, SD is 1.0
The metric problem is when you’re trying to compare an unknown metric. Translates it to the score that we are familiar with.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Given a z-score for a person, describe in words their relationship to the rest of the distribution (that is, how far above or below are they from the mean in SD units).

A

Example:
1.23 standard deviations above the mean
-1.23 is 1.23 standard deviations below the mean.
Z scores tell you how above or below the mean the scores are.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

level of significance/alpha

A

the point at which the p value is compared to reject the null hypothesis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

statistical significance

A

when you reject the null hypothesis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Type 1 error

A

When you reject the null and you should not

false positive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Type 2 error

A

When you fail to reject the null and you should

false negative

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

power

A

he ability to reject the null when the null is false.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q
  1. Describe when a statistic is said to be “statistically significant” and what it means for a statistic to be statistically significant.”
A

If p value is lower or equal to the alpha level it is considered to be statistically significant.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

List one way to increase the power of a statistical test.

A

Increase sample size

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

List and describe the two features of the correlation coefficient.

A

Direction - Positive and negative

Strength - 0 to 1, strong, moderate, weak

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Given a correlation describe the nature of the relationship between the two variables.

A

direction
strength
signifiance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Describe the ideas behind the partial correlations coefficient (“What does it mean when something is being ‘controlled for’?)

A

Partial correlation coefficient is looking for a relationship between two variables when removing the influence of the third variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the regression equation used for?

A

Regression allows you to make a prediction. Predict Y from values of x

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Recall the formula for a simple regression line (the “line of best fit”) and label each part.

A

o Line of best fit is the line that minimizes the error or residual
o Formula for a straight line Y^1=b1X +b0
b1= slope of the line, b0= y intercept

X= IV and Y= DV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Describe how the beta weights are similar to partial correlation coefficients.

A

The IVs are highly correlated with each other.
BOTH represent the relationship between an independent variable and a dependent variable, after removing the influence of all other independent variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Describe the purpose of factor analysis.

A

To simplify our data set and not have as many variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Describe how factor analysis aids in exploring the dimensionality of a scale/test.

A

You can compare with a factor analysis the number of dimensions you think should exist in your scale with the number of factors that actually emerge in the analysis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

factor:

A

A group of intercorrelated items

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

eigenvalue:

Study These Flashcards
A

The eigenvalue is the variance the factor is accounting for

23
Q

scree curve/plot

Study These Flashcards
A

The Scree plots is the plot of the factor against their respective eigenvalues.

24
Q

communalities

Study These Flashcards
A

Communalities of the variables are the proportion of variance each variable shares with the entire factor solution

25
factor loading:
Factor loadings are the correlations between the respective items and the factors
26
marker variables
Marker variables are the items with the highest correlation with each factors
27
simple structure
strong correlation with one and a low correlation with other factors
28
split factor loading
split factor loadings is when a variable is equally correlated with two or more factors
29
Given a mean and a standard deviation for a distribution, describe the nature of the distribution (that is, where is approximately 68% or 96% of the distribution located?).
68 percent of the distribution is located between -1 SD and 1SD. This is considered the average range. 96 percent of the distribution is between the -2 SD and 2 SD. This encompasses all except for the top 2% and the bottom 2%.
30
Meta Analysis
Quantitative review of research literature | a study of studies
31
Define effect size and why effect sizes are used in meta-analysis.
It’s a standardized measure of a treatment effect. | Why→ The measure of the treatment becomes standardized when using effect sizes allowing us to compare studies.
32
Give the formula for d
3. d= Mean Control Group - Mean Treatment Group/ SD of the control group. (It can either be the SD of the control group or the pool SD)
33
Given effect sizes from some studies calculate the overall/grand effect size.
Σd/N | Sum of all individual effect sizes divided by the number of studies (number of effect sizes)
34
independent samples/between-groups design
In between group design, separate and distinct groups are compared. Examples: Most experimental designs compare men to women, treatment group compared to control, African vs. Hispanics, MFT vs, General Psychology
35
related samples/within-groups design
Within Group Design is when one group is tested repeatedly. 
Examples: Group paired to itself, pre-test vs. post-test
36
factor
Each independent variable
37
level
Each subgroup of an independent variable
38
one-way ANOVA
The one-way analysis of variance (ANOVA) is used to determine whether there are any statistically significant differences between the means of two or more independent (unrelated) groups (although you tend to only see it used when there is a minimum of three, rather than two groups). 1 DV and 1 IV
39
between group variance
Between-group/across group variance is explained by the Independent Variable.
40
post hoc analyses
When there’s a significant ANOVA, three or more levels and we look at a significant difference between every pair of means.
41
F ratio
The ratio of variance F test: Between Group Variance/Within Group Variance In other words, it’s the variance due to the IV divided by the variance that’s not due to the IV.
42
factorial ANOVA
2+ IVs and 1 DV.  You do a factorial ANOVA when you think there’s a moderating effect.
43
main effect
Also known as the direct/simple effect. The impact of an IV on the DV (ignoring/independent of all other IV’s).
44
marginal mean
The marginal means are the means for the levels of one IV.
45
Interaction
The combined impact the IVs are having on each other’s relationship with the DV.
46
MANOVA
One IV and 2+ DVs. Whenever we run tests with multiple DVs you run the risk of making Type 1 Error. A MANOVA tests for significant differences across levels of the IV for all DVs simultaneously (as a group/set).
47
Describe the logic behind the calculation of the F-ratio. Plus, if parts of the ANOVA printout were missing be able to fill in blanks using the information provided.
* 1st part→  The ANOVA table is read from left to right. * 2nd part→ The total variance is split:  between variance/ within variance. Divide it by Degrees Of Freedom (get the mean square) * 3rd part → Recombine by then dividing the (mean square) between variance by (mean square)within variance to get the F ratio.
48
Discuss when and why post hoc contrasts are used in ANOVA.
When looking at the overall significance level of the F-test (ANOVA), if it is statistically significant meaning p-value is less than .05 then you run a Post hoc test to determine if there’s a significant difference between every pair of means or only within some of them to determine which ones are actually affected by the IVs.
49
Given a verbal description of a factorial design, state what kind of design it is (like “a 2 × 3 factorial design”), be able to list how many factors there are, how may levels in each factor and how many cells are involved in the design.
a 2x3 factorial design would be there are 2 factors (IVs) with 2 levels, 3 levels (subgroups), 6 Cells a 2x2x3 factorial design would be 3 factors (IVs) with 2 levels X 2 levels X 3 levels (subgroups), 12 cells
50
Discuss what it means that main effects are qualified by the presence of a significant interaction.  Given an illustration.
When there is no significant interaction the main effects can be interpreted as a simple effect. Significant interaction takes precedence over main effects (if there is an interaction it will affect your understanding of the main effect, the main effect is not telling you the whole story).
51
Describe the research situation when a MANOVA would be used (in contrast to an ANOVA).
You run a Manova when you have multiple(min.2) DVs and one IV. You run an Anova when you have multiple IVs and one DV
52
Describe how MANOVA is considered by some to control for Type I error (that is, why not just run multiple ANOVAs?).
Whenever we run tests with multiple DVs you run the risk of making Type 1 Error (increases your chances of getting a false positive) → each chicken runs across the street one at a time. A MANOVA tests for significant differences across levels of the IV for all DVs simultaneously (as a group/set) → all the chicken run across the street together.
53
Describe the interpretation of the Lambda statistic.
* It goes from 0 to 1 | * Lambda is the proportion of variance the IV does NOT explain of the DV
54
Describe how a significant lambda is to be followed up/explored
A significant lambda is followed up with univariate follow up tests (i.e. 20 individual ANOVAs)

stats (1 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions