descriptive and inferential statistics

Question 1

What are examples of forms of raw data?

Answer 1

a. Questionnaires
b. Interviews
c. Objective measures

Question 2

Explain descriptive vs inferential statistics

Answer 2

Descriptive statistics
These are basics measures
They average scores on variables and how they differ

iii. Cannot conclude statistical differences

Inferential statistics
These help make decisions about the Null and research hypotheses
They make Generalizations from the sample population

Question 3

What is central tendency

Answer 3

used for summarizing large amounts of data into one typical or average value

Uses mean median and mode

A summary of measure that attempts to describe a whole set of data with a singe value that represents the middle or centre of its distribution

Question 4

Explain what the mean its level of measurement and when it is used

Answer 4

Mean measures the average score for each variable.
- Level of measurement = interval and ratio
- Used when: you can, and the data fit

Question 5

Explain what the median its level of measurement and when it is used

Answer 5

Median is the mid point of data

a. Level of measurement: ordinal
b. Used when: data include extreme scores

Question 6

Explain what mode is, its level of measurement and when it is used

Answer 6

Mode is the most frequent score

a. Level of measurement: nominal
b. Used when: data are categorical

Question 7

What are the indices of variability?

Answer 7

a. Range
b. Standard deviation
c. Skewness

Question 8

What is the importance of descriptives?

Answer 8

Provide a first glance at the data
i. Describes the variability of the sample
b. Allows other researchers to understand the data

Question 9

9. All variables are assumed to follow a ____ distribution

Answer 9

normal

Question 10

What is the meaning of z-scores?

Answer 10

the number of standard deviations a particular score is away from the sample mean

a. Different scores represent different locations on the x-axis
i. The location on the axis is associated with a certain %

b. Z-scores can be used to predict:
i. Percentage of scores both above and below a particular score
ii. The probability that a particular score will occur in a distribution

c. Z-scores are a good method to understand how scores relate to one another
i. Typically used at population level, but can be useful in samples however not always practical.

if you want to compare it to the “average” person’s weight, looking at a vast table of data can be overwhelming (especially if some weights are recorded in kilograms). A z-score can tell you where that person’s weight is compared to the average population’s mean weight
…
let’s say you have a test score of 190. The test has a mean (μ) of 150 and a standard deviation (σ) of 25. Assuming a normal distribution, your z score would be:

z = (x – μ) / σ
= (190 – 150) / 25 = 1.6.
The z score tells you how many standard deviations from the mean your score is. In this example, your score is 1.6 standard deviations above the mean

way to compare results to a “normal” population

Question 11

If a z-score has a p value less than 0.05, it is deemed _ _ than the comparison

Answer 11

Significantly different
Reject the null
Want it to be less than .05

Question 12

What is a type 1 error?

Answer 12

Error of optimism, resulting from rejecting the null hypothesis when it is true.

a. The probability of making a type 1 error:
- Set by researcher
(Ex. 0.01% chance of rejecting the null when it is true)

Question 13

What is a type 2 error and how can risk of this be reduced?

Answer 13

Error of pessimism, resulting from accepting the null when it should be rejected

• Reduced by increasing sample size
• Reduced by increasing treatment effects

Question 14

List the 8 steps of using a significance test:

Answer 14

a. State the null hypothesis

b. Establish the significance level

c. Select the appropriate test statistic

d. Compute the test statistic (obtained value)

e. Determine the value needed to reject the null (critical value)
i. This depends on:
1. Level of significance chosen (e.g. p = 0.5)
2. Degrees of freedom (based on sample size)

f. Compare the obtained value to the critical value

g. If the obtained value is greater than the critical value, reject the null

h. If the obtained value is less than or equal to the critical value, accept the null

Question 15

Explain non-parametric
(Remember the measures)

Answer 15

a. Data are nominal or ordinal
b. Population is not normal
c. Useful in frequency data or proportions

Tests:
Chi square
Mann Whitney wilcoxan u (compares 2 samples or groups)

Question 16

Explain parametric data

Answer 16

Data are interval or ratio level

Population from which observations are made are normally distributed

Tests:
One sample t-test
Independent sample t-test
Dependent sample t-test

Question 17

What are the 3 types of t-tests?

Answer 17

a. One sample t-test
b. Independent sample t-test
c. Dependent sample t-test (paired t-test)

Used for parametric data

Question 18

Explain one sample t-test

Answer 18

Compare sample mean to known population mean

Only need 1 sample of collected data but need to know mean you are comparing against

Don’t need to know population standard deviation; use info collected from sample

Question 19

Explain independent sample t-test

Answer 19

a. Comparing results of 2 groups and the Individuals in each group are different

Ex. Male and female

Question 20

Explain dependent sample t-test

paired sample

Answer 20

Comparing 2 means but individuals in the 2 samples are the same people

need to account for person effects within the calculation, hence the different test

Question 21

Explain the difference between a 1-tail and 2-tail test:

Answer 21

2 tailed
- p = 0.05, the critical z =1.96

1 tailed
- p =0.05, critical z = 1.65

knowing the direction of a result is needed ot use a 1-tailed test

1 tailed tests are more lenient (will reject the null more) then 2 tailed tests

with 2 tailed tests, it is assumed we reject the null unless falls within critical regions

Question 22

what types of analysis are used for experimental studies?

Answer 22

t-test
ANOVA (multiple types)
chi-square

Question 23

what type of analysis is used for correlational studies?

Answer 23

bivariate correlations
regression analyses (many types)

Question 24

what types of analysis are used for multivariate designs

Answer 24

MANOVA
hierarchial linear model

Question 25

how is a z score (standard score) calculated?

Answer 25

z = individual score - mean of the group / population standard deviation

Question 26

if the null hypothesis should be accepted and you rejected it, this is a ____ error

Answer 26

type 1

Question 27

if the null hypothesis is false and you accept it, this is a ____ error

Answer 27

type 2

Question 28

why should we beware large samples?

Answer 28

effects can be statistically significant but not clinically important

Question 29

why should we beware the small sample?

Answer 29

it is hard to find significant differences
& no difference means little or nothing

Question 30

what is the F-test

Answer 30

Used for tests with multiple groups

the f-test will tell us a significant difference exists somewhere between the groups but will not tell us where
- so you can reach the conclusion that significant difference does exist but need to follow up with other test to find out where

Question 31

steps in f-test follow up

Answer 31

• post-hoc analyses (multiple types exist) - pinpoints where the difference exists
• allows to investigate difference once significant test has been found
• adjust the alpha value due to EWE
• pinpoint where differences exist
• can then reach final conclusions about groups

Question 32

2-way ANOVA

Answer 32

Investigates the interaction between independent variables 1 and 2

Question 33

1-way ANOVA

Answer 33

investigate tests of 3 or more population means with just 1 independent variable

Question 34

If the obtained value is greater than the critical value…

Answer 34

Reject the null

Question 35

If the obtained value is less than the critical value…

Answer 35

Accept the null

Question 36

EWE meaning

Answer 36

Experiment wise error