Topic 4

Question 1

How do scientists decide what to measure?

Answer 1

1. Based off of other experiments
2. Modifying common,y used measures
3. Refining the constructs of interests (operational definitions)

Question 2

Define habituation

Answer 2

A gradual decrease in responding to repeated stimuli

Infants will look at new stimuli over old stimuli

Question 3

Define reliability

Answer 3

Results are repeatable when behaviours are remeasured

Minimal measurement error

Question 4

Define validity

Answer 4

If the measurement measures what it is supposed to, it is valid.

Also determines if the experiment was properly conducted and if the hypothesis has been tested

Question 5

Levels of validity

Answer 5

1. Content validity - content on a test is related to the construct being measured
2. Face validity - if the measure seems valid to participants
3. Criterion validity - if the measure can accurately predict future behaviour or is meaningfully related to some other behaviour
4. Construct validity - whether a test adequately measures a construct
   - convergent/discriminant validity measures how related results are to other experiments measuring the same construct

Question 6

____ assumes _____ but ____ does not mean an experiment is ______

Answer 6

Validity, reliability, reliability, validity

Question 7

Scales of measurement

Answer 7

1. Nominal scales - studies that assign people to categories and count the amount of people in each group
2. Ordinal scale - sets of rankings to show the relative standing of objects/individuals
3. Interval scales - 0 does not mean nothing
4. Ratio scale - 0 is the absence of the concept being measured

Question 8

Define population

Answer 8

All members of a group

Question 9

Define sample

Answer 9

A subset of the population group

Question 10

Descriptive statistics

Answer 10

• summary of the data collected from the sample of participants
• mean: average of all data
• median: score in the middle of a set
• mode: score that occurs the most frequently
• range: difference between highest and lowest score
• standard deviation: an estimate of the average and how much the same scores deviate from the mean
• variance: number in standard deviation before taking the square root

Question 11

How to calculate standard deviation

Answer 11

1. Calculate the mean
2. Subtract the mean from each score
3. Square each value
4. Divide the sum of squared values by the sample size - 1
5. Take the square root

Question 12

Inferential statistics

Answer 12

• draw to conclusions about data that can be applied to wider populations

Question 13

Null hypothesis

Answer 13

Assuming there is no difference in performance between conditions being tested
Fail to reject - any difference in means is likely due to chance
Reject - a cause and effect relationship is noticed and results can be generalized

Question 14

Alternative hypothesis

Answer 14

The outcome you are hoping to find in an experiment

Question 15

Type 1 vs Type. 2 error

Answer 15

Type 1
- reject null when it is true
- support alternative when it is false
- thinking there is a significant relation when there is none
Type 2
- fail to reject null
- fail to support alternative
- you don’t find a significant relation even though there is one

Question 16

Systematic variance

Answer 16

The result of an identifiable factor being inadequately controlled

Question 17

Error variance

Answer 17

Individual differences within and between groups and any random unpredictable effects that may have occurred during the study

Question 18

Inferential statistic formula

Answer 18

Variability between conditions (systematic and error)/ variability within condition (error)

Question 19

Interpreting failures to reject null hypothesis

Answer 19

• done with caution as it is easy to miss 1 difference
• repeat experiment to prove no differences in groups
• no significant findings are not published (file drawer effect)

Question 20

Effect size

Answer 20

Provides an estimate of the magnitude of differences among sets of scores while accounting for the variability in scores
- allows researchers to combine results from diverse experiments with different operational definitions of the same construct

Question 21

Confidence interval

Answer 21

An inferential statistic that allows researchers to draw conclusions of the whole population based on sample data
-range of value including a population value with a certain degree of confidence
- no overlap = significant difference

Question 22

Power

Answer 22

The chance of being able to reject a null hypothesis
- as power increases, type 2 errors decrease
- affected by alpha level, effect size, and sample size