13 - Other

Question 1

What can we conclude if a b-value has a large standard error? (Other Statistics)

Answer 1

The estimates of b vary widely across different samples (estimates could be very different from the population value)

Question 2

What does Rsquared tell us? (Other Statistics)

Answer 2

How much variation is accounted for (x100 to give a %)

Question 3

What do confidence intervals tell us? (Other Statistics)

Answer 3

The interval is one of the 95% that contains the population value of b lies between x and x

Question 4

What does the F statistic represent? (Other Statistics)

Answer 4

The ratio of variance in the variable, explained by the model to the variance that is not explained by the model

Question 5

What does bootstrapping assume? (Other Statistics)

Answer 5

Nothing about the shape of the sampling distribution

Question 6

What does sphericity refer to? (Other Statistics)

Answer 6

The equality of variances of the difference between treatment levels

Question 7

What is the equation for odds? (Other Statistics)

Answer 7

Odds = no of times an even occurs / no of times an event does not occur

Question 8

What are the predictor and outcome variable respectively known as? (Other Statistics)

Answer 8

Independent and dependent variable

Question 9

What does it mean if the values of Rsquared and adjusted Rsquared are close together? (Other Statistics)

Answer 9

The model should generalise well to the population

Question 10

What does the beta estimates for a predictor represent? (Other Statistics)

Answer 10

The change in outcome for each unit change in the predictor

Question 11

What does the F and P values tell us in a linear model with multiple predictors? (Other Statistics)

Answer 11

If the overall model explains more variance in the outcome than it does not explain

Question 12

What does a P value represent? (Other Statistics)

Answer 12

The probability that you would get a value at least as extreme as the observed, if there is no real relationship between the predictor and the outcome (the null hypothesis is true)

Question 13

What does a t-test for the predictor value do? (Other Statistics)

Answer 13

Tests if the beta value differs significantly to zero

Question 14

What is the difference between the F and T value? (Other Statistics)

Answer 14

F value = whole model improved against the mean

T value = Each predictor individually

Question 15

What is the difference between regression and ANOVA? (Other Statistics)

Answer 15

Nothing

Question 16

What does repeat testing groups from the data set do? (Other Statistics)

Answer 16

Increases the chances of type 1 error

Question 17

What is a type 1 error? (Other Statistics)

Answer 17

• Reporting a significant finding, where there is no real effect
• Too lenient

Question 18

What is a type 2 error? (Other Statistics)

Answer 18

• Reporting no significant finding, where there is one

- Too stringent

Question 19

What is the maximum number of contrasts within x groups? (Other Statistics)

Answer 19

x - 1

Question 20

What does it mean if there is a significant interaction between predictor and covariate? (Other Statistics)

Answer 20

Homogeneity of slopes has been violated

Question 21

What does a covariate in ANCOVA do? (Other Statistics)

Answer 21

Adjusts the means in each group to control for a known confound

Question 22

Sphericity is and assumption of what and what does is assume? (Other Statistics)

Answer 22

Is an assumption of repeated measures and assumes that the variances of the differences between conditions are equal