Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

CHYS 3P15 > After Midterm > Flashcards

After Midterm Flashcards

1
Q

How can we predict a value from a second value?

A

Through Bivariate Regressions

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

What does Bivariate mean?

A

It means one is predicting the other

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

What is a Regression? (Or Linear Model)

A

It is a way of predicting the value of one variable from another

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

What equation do you use for a regression

A

The equation of a straight line

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

What variables are in a regression

A

1 IV and 1 DV
We create this model and we decide based on theory which will be IV and which with be DV

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

What assumptions are in a regression

A

Normality
Linearity
Homogeneity of Variance
Homoscedasticity
Outliers

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

What is the slope? (b1)

A

How steep or flat is the line.
If it is flat then there is no association or it’s just 0

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

What is the intercept? (b0)

A

Starting point of the DV before you even predict it
(It’s like what are our depression symptoms without any starting point)

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

For a correlation what variables are on which axis

A

It does not matter

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

For a regression which variable is on which axis

A

IV needs to be on X
And DV needs to be on Y

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is b1 called?

A

Regression coefficient
It can be positive or negative
Tells us direction (positive negative)
And strength/magnitude

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

When do you decide if you are doing a correlation or a bivariate regression?

A

In the research question, if it says predict then you are doing a regression
If it says association then you are doing a correlation

Both of these require 2 continuous variables

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

When do you want to see an association of X and Y?

A

In a correlation

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

When do we predict Y from X

A

In a bivariate regression

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

Why do we predict scores if we already know it and it’s a variable in our data set for Bivariate regressions

A

We want to see how accurate our model is to the real data we collected

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

What are residuals in a bivariate regression model?

A

Difference between our predicted Y and actual observed Y
So if our model predicts 6 and we had a score of 8 the residual is 2

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

Can residuals be positive or negative

A

Yes they can be both

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

True or false: the smaller the deviation or “error” the better our model “fits” or represents our data

A

True

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

How can deviations be calculated

A

Deviance = outcome - model

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

What is SSt

A

Total variability (variability between scores and the mean)

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

What is SSr

A

Residual/error variability (variability between the regression model and the actual data)

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

What is SSm

A

Model variability (difference in variability between the model and the mean)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

True or false: if the model results in better prediction than using the mean, then we expect SSm to be much greater then SSr

A

True

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What does the ANOVA tell us

A

Tells us how well our model fits the data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
What is the ANOVA test/ statistic
It is the F test It looks at whether the variance explained by the model (SSm) is significantly greater than the error within the mode (SSr)
26
What order do we interpret the ANOVA
ANOVA then the model summary then the coefficients
27
What does the ANOVA explain
How good the model is vs how bad
28
What is the alternative and null hypotheses for ANOVA
Null: the model does not fit our data well (example: sleeping will not predict depression symptoms well) Alternate: the model fits our data well (Sleeping will predict depression symptoms well)
29
What is the R2 in ANOVA
The proportion of variance accounted for by the regression model
30
What is the effect size for R2
.02 small .13 medium .26 large
31
In ANOVA, if p < .001 do we reject or fail to reject the null
We reject the null and it means our model is fitting the data well
32
What’s something important about the ANOVA table
Our sum of squares needs to get divided by the degrees of freedom to get the mean square
33
Which value allows for some comparison across scales and other studies
The beta value
34
When do you interpret the slope
You only interpret the slope when the p value is significant (less then .05)
35
What does a multiple regression do?
Predicts a value with several additional values There are multiple predictors (IV) And one outcome (DV)
36
How many continuous variables does a multiple regression have
Several But still only one DV (outcome)
37
What equation does multiple regression use?
It uses the equation of a line
38
What assumptions does multiple regression have
Normality Linearity Homogeneity of variance (homoscedasticity) Outliers
39
What is outcome = model + error
Outcome is DV Model is IV
40
If there is 1 predictor are b and r the same?
Yes
41
What is b0 in a MR
The intercept Point at which the regression line crosses the y axis
42
What is b1 in a MR
Regression coefficient for predictor x1 accounting for X2 Slope of the line and direction/strength
43
What is b2 in MR
Regression coefficient for predictor X2 accounting for X1 Slope of the line and direction/strength
44
How do we estimate the model in MR
The method of least squares (the smaller the error the better the model)
45
What is how much we can explain and what is how much we cannot explain?
How much we can: Mean Square How much we cannot: Residual
46
Unstandardized coefficient is in what scale
Your original scale
47
Standardized coefficient is in what scale
It is in standardized deviation units
48
What is a forced entry (simultaneous regression) method of regression
All predictors are entered simultaneously
49
What is a hierarchical method of regression
You decide (the researcher) which IV go in, and you decide in what order This is the best method when based on theory testing but the researcher needs to know what they’re doing
50
Difference between bivariate and multiple regressions
In BR we predict Y from X In MR we predict Y from X1, X2, X3
51
What does this equation mean F= MSm/MSr
How much are you able to explain over how much you are not able to explain This is based on your total variance MS= Mean squares
52
What is the Pearson correlation coefficient squared?
It is the R2 value It is the proportion of variance accounted for by the regression model
53
What are the R2 value effect sizes
.02 small .13 medium .26 large
54
How can you tell if it is simultaneous or a hierarchical regression
Look at the writing underneath the ANOVA box It will tell you if all the IVs were added at the same time simultaneously
55
For an MR if we reject the null what does it mean
It means our variables are predicting what they’re intending to predict (It is significant)
56
What do we convert R2 to?
We convert it to a percentage It tells us how much our IVs are explaining our outcome For example .523 means our 3 IVs together are explaining 52.3% in depressive symptoms
57
Does every b value have its own p value?
Yes
58
In the coefficients table.. do we only interpret the B values that have an sig value of less then .05?
Yes And it would be interpreted like this: “for every 1 unit increase in perpetration we are seeing depression increase by 1.24 units (b value)”
59
What is a partial correlation in bivariate correlations?
Controlling for a third variables association to both variables
60
What is a semi partial correlation in multiple regression?
It measures the relationship between a predictor and the outcome accounting for the relationship between that predictor and any others already in the model -it measures the unique contribution of a predictor to explaining the variance of the outcome
61
Where is the semi partial correlation on an SPSS output
It is in the coefficient table under “part” We want to square that number (times it by itself) to get it to a percentage This gives us the percentage of overlap ONLY for significant predictors tho
62
What is the last step is Multiple regressions?
Calculating the output coefficient “All 3 were collectively explaining 52.3% of depression symptoms. Perp is explaining 4.8%, victimization 39.3% and sex is explaining 0%”
63
What words do you look for when determining if a hierarchical regression is being used?
The research question or the null and alternate will have the words AFTER ACCOUNTING FOR
64
What do the 2 steps indicate in a hierarchical regression?
Step 1 and step 2 have their own F ratio Step 1 needs to be interpreted before step 2 We are interested in step 2 because that is specific to the hypothesis we are testing
65
What is R2 change and F change for hierarchical regression?
We want to know does the new model predict better than the old model? Is it a significant improvement?
66
How to get what the total model accounted for in R square change for hierarchical regression
Add step 1 r square change and step 2 r square change Example .130 + .393 =0.523 which means total model accounted for 52.3% of variance 13% to one variable and 39.3% to the other
67
True or false: you must interpret the b coefficient in the step that the predictor was first entered in
True If 2 variables are in step 1 And 3 are in step 2 You interpret the first 2 in step 1 and the last one only in step 2
68
What step do you construct the regression equation
Since step 2 is your final model you do it all from step 2
69
What is an independent ANOVA
There is a control (no school intervention) Experimental group 1 (bullying intervention) Experimental group 2 (social skills intervention)
70
What variables does an independent ANOVA have
It has 1 continuous and 1 categorical But it has at least 3 or more groups
71
What does the independent ANOVA do
It compares each combination of groups And compares to a base group (usually the control group)
72
What does the ANOVA f statistic tell us
It tells us if the experiment is successful then the model will explain more variance then it can’t (SSm will be greater than SSr)
73
What do planned contrasts and post hoc follow up tests do?
When we know there is a group difference in our model we use one of these tests to find where
74
Rules when choosing contrasts
Independent (must test unique hypotheses) Only 2 chunks (each contrast should only compare 2 chunks of variation) K-1 (you should always end up with one less contrast than the number of groups. So with 3 group there are only 2 comparisons)
75
How do you choose contrasts?
When there is usually one or more control groups The first contrast will always be to compare any control groups (chunk 1) with any experimental groups (chunk 2)
76
Why do we need to interpret the assume equal variances in contrast tests?
We need to see if there is a significant value. If there is we continue interpreting the top row
77
Which Post-Hoc test do you do
Do bonferroni if you meet your equal variances Do dunnet if you violate it
78
If you violate homogeneity in your post hoc test which statistic do you interpret
You interpret Welch NOT F ratio
79
What are the effect sizes for ANOVA independent test
.2 small .5 medium .8 large
80
What is a two way independent ANOVA (factorial ANOVA)
It is when there is 2 IV with groups Example: Puppy therapy (15 mins or 30 mins) And Student type (elementary or high school)
81
What are the variables in a two way ANOVA
1 continuous (DV) 2 categorical (IVs)
82
Does a factorial ANOVA has different participants in all conditions?
Yes This is also a between-subjects design
83
What is a factorial ANOVA trying to see
If there is group based differences
84
When we see the word “interaction” we know it is a
Factorial ANOVA
85
How do we know if there is a significant interaction in factorial ANOVA
If there is an overlap on the graph We do not know if that is a significant difference though until we look at our p value

CHYS 3P15 (4 decks)

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