Week 11

Question 1

What kind of data is used in parametric tests?

Answer 1

Numerical data (interval or ratio).

Question 2

What kind of data is used in non-parametric tests?

Answer 2

Categorical data (nominal or ordinal).

Question 3

Parametric tests normally make assumptions, what are they?

Answer 3

Population parameters such as mean and normality of distribution.

Question 4

List the 3 main reasons for using non-parametric tests.

Answer 4

1. Numerical scores violate assumptions of mean and normality of distribution (unless solved with sample information).
2. High level of variance in numerical data (converting numerical to ordinal data)
3. Grouping is sometimes more convenient (scores, participants etc).

Question 5

What does a Chi-square goodness-to-fit test measure?

Answer 5

Uses sample data to determine how well a sample proportions fit the population proportions based on the null hypothesis. (How well ƒₒ fits with ƒₑ)

Question 6

What are the TWO hypothesises for a Chi-square goodness-to-fit test.

Answer 6

1. Equal proportions/No preference (50/50 or 33.33/33.33/33.33 etc).
2. No difference (to population proportions).

Question 7

Is there a alternative hypothesis for a Chi-square goodness-to-fit test?

Answer 7

No.

Question 8

What is observed frequency in a Chi-square goodness-to-fit test?

Answer 8

The data observed in the sample.

Question 9

What is expected frequency in a Chi-square goodness-to-fit test?

Answer 9

What the distribution is expected to look like.

Question 10

What symbol is used to represent expected frequency in Chi-squared tests?

Answer 10

ƒₑ

Question 11

What symbol is used to represent observed frequency in Chi- squared tests?

Answer 11

ƒₒ

Question 12

What symbol is used to represent Chi-squared?

Answer 12

𝜒²

Question 13

What is the formula for Chi-squared?

Answer 13

𝜒² = ∑(ƒₒ-ƒₑ)² / ƒₑ

Question 14

What is Chi-squared?

Answer 14

A theoretical, positively skewed distribution consisting of all possible random samples that would exist if the null hypothesis is true.
(Similar to a t-test distribution).

Question 15

What does the Chi-squared of independence test measure?

Answer 15

Determines if a relationship exists between two categorical variables.

Question 16

What is the null hypothesis for a Chi-squared of independence test?

Answer 16

X is NOT an independent variable for Y.

Question 17

What is the alternative hypothesis for a Chi-squared of independence test?

Answer 17

X IS an independent variable for Y.

There is a relationship between the variables

Question 18

How do you calculate ƒₑ for a Chi-squared of independence test?

Answer 18

ƒₑ = ƒ꜀ƒᵣ/n

Question 19

What does ƒ꜀ represent?

Answer 19

Column frequency.

Question 20

What does ƒᵣ represent?

Answer 20

Row frequency.

Question 21

How do you calculate ƒₑ for a Chi-squared goodness to fit test?

Answer 21

ƒₑ = pn

proportion times sample

Question 22

How do you calculate the df in a Chi-squared of independence tests?

Answer 22

df = (R-1)(C-1)

(Row sample size - 1 times column sample size -1

Question 23

What is risk in statistics?

Answer 23

The probability that an event occurs.

Question 24

What is a basic calculation of risk in words? (hint: it is the equation for probability).

Answer 24

Number of outcomes classified as A divided by the total number of outcomes.

Question 25

What is odds in statistic?

Answer 25

The number of times that the event will occur divided by the number of times that the event will not occur.

Question 26

What is relative risk in statistics?

Answer 26

Comparing the risk of the same event between two groups.

Question 27

What is the calculation of relative risk in words?

Answer 27

Risk that the event occurs for group 1 divided by risk that the event occurs for group 2.

Question 28

What is the calculation of an odds ratio in words?

Answer 28

Odds of an outcome for group 1 divided by odds of an outcome for group 2.

Question 29

How to do you preform relative risk and odds ratios in SPSS?

Answer 29

1. Analyse.
2. Descriptive statistics.
3. Crosstabs.

Question 30

How do you calculate the df in a Chi-squared goodness to fit tests?

Answer 30

df = number of categories - 1

Question 31

What assumptions do nonparametric tests make?

Answer 31

1. Do not require the populations to have normal distributions.
2. Can be performed on data measured on any scale.
3. Make few, if any assumptions about the populations.

Question 32

How does the difference between ƒₒ and ƒₑ influence the outcome of a chi-square test?

Answer 32

​The larger the difference, the larger the value of chi-square, and the greater the likelihood of rejecting the null hypothesis.

Question 33

What happens to the critical value for a chi-square test if the size of the sample is increased?

Answer 33

The critical value depends on the number of categories, not the sample size.

Question 34

For a fixed level of significance, the critical value for the chi-square will increase as the ____ increases?

Answer 34

Degree of freedom.