Statistical Models

Question 1

A statistical model gives a simple

Answer 1

description of relationships in the dataset

Question 2

A statistical model uses

Answer 2

maths to summarise a dataset relative to multiple variables.

Question 3

——— statistics use statistical models

Answer 3

inferential

Question 4

Correlation is an example of a

Answer 4

statistical model

Question 5

Data = model + ……….

Answer 5

error

Question 6

error is sometimes called

Answer 6

residuals

Question 7

The smaller the error, the better the model —–

Answer 7

“fit”.

Question 8

A z-score describes

Answer 8

the position of a raw score in terms of its distance from the mean, when measured in standard deviation units.

Question 9

The z-score is —— if the value lies above the mean, and —— if it lies below the mean. A z-score of 1 = 1SD away from the mean.

Answer 9

positive, negative

Question 10

z-scores tell how an

Answer 10

individual sits within a distribution.

Question 11

whats the difference between a hsitogram and a density plot

Answer 11

a histogram shows the counts of values in each range, while a density plot shows the proportion of values in each range.

Question 12

The p value is a measure of

Answer 12

probability

IT IS A DECISION MAKING TOOL

its NOT
-probability that null hypothess is true,
-the probability you are making the wrong deciion
-It is not the probably that if you ran the study again, you would obtain the same result that % of the time.
-It does not mean you found an important effect.
-It does not reflect the size of the effect.

Question 13

Pvalue is a ——– assessment

Answer 13

Dichotomous assessment.

Question 14

Pvalue is a dependant on

Answer 14

Dependent on your alpha value and your p value.

Question 15

The alpha is typically set at

Answer 15

0.05 meaning that p<.05 is deemed statistically significant.

Question 16

In setting your alpha value you are balancing

Answer 16

Type I and Type II errors.

Question 17

Effect size

Answer 17

Measure of the strength of the effect.

Question 18

There are different ways to measure effect size,

Answer 18

e.g. r2, Cohen’s d.

Question 19

A statistical hypothesis test is a test of

Answer 19

the statistical hypothesis, not the research hypothesis.

Question 20

the goal behind statistical hypothesis testing is not to eliminate errors, but to

Answer 20

minimise them.

Question 21

. A statistical test is pretty much the same. The single most important design principle of the test is to control the probability of a type I error, to keep it below some fixed probability. This probability is called the

Answer 21

significance level of the test

Question 22

A “powerful” hypothesis test is one that has a small value of ——, while still keeping —- fixed at some (small) desired level.

Answer 22

beta, alpha

Question 23

critical values since they define the

Answer 23

edges of the critical region

Question 24

If the data allow us to reject the null hypothesis, we say that “the result is

Answer 24

statistically significant

Question 25

Significant in the context of statistics does not mean

Answer 25

important, it rather means indicates

Question 26

All that “statistically significant” means is that the data allowed us to

Answer 26

reject a null hypothesis.