week 6

Question 1

What is correlational design

Answer 1

We look at pairs of scores to see whether scores on one measure are consistently associated with scores on another measure
We measure both variable for each person in our sample

Question 2

Correlational analysis

Answer 2

Correlation result have components: Visual pattern of relationships- descriptive. Numerical description of relationship- is inferential

Steps: are the data suitable, visualise the data to see the relationship, choose type of correlation, run correlation results, interpret the correlation results, report the correlation results

Question 3

Level of measurement

Answer 3

Categorical, Ordinal, scale
categorical-> use chi squared analysis, not correlation
Ordinal, scale-> correlation

Question 4

Scale data

Answer 4

equal intervals means that the unit of difference between adjacent points on the scale is the same, regardless of where they are on the scale

Question 5

Ordinal data

Answer 5

Categories that can be ordered- property of magnitude, but no precise difference between ranks, so the categories have an order, they might not be evenly spaced

Question 6

visualising data with scatter plot

Answer 6

Dots show scores on both variables; each dot represents an individual
One variable on the X axis, the other on the Y axis; it doesn’t ,atter which one is on which axis
A scatterplot can give us information

Question 7

Linear relationships

Answer 7

a relationship between two variables that can be describe by a straight line. The stronger the positive or negative, the less the various point will depart from the straight line

Question 8

positive relationships

Answer 8

as one variable increases so does the other variable an uphill line

Question 9

Negative relationships

Answer 9

as one variable increases the other variable decreases

Question 10

spot restricted ranges

Answer 10

when range is restricted, correlation can go down

Question 11

Subgroups

Answer 11

scatterplots can show us whether we have a false positive correlation due to different subgroups of participants

Question 12

related variables

Answer 12

only refer to variables being related when describing results of correlational studies

Question 13

Pearson’s product-moment correlation

Answer 13

this is a parametric test so it’s the most powerful correlation
Assumptions: type- data should be continuous, rather than ordinal
Normal- both variables should approximate a normal distribution
Extreme- there should be no extreme values, outliers can overly influence that calculation of Pearson’s statistic, more so than the other data points

Question 14

Spearman’s Rank correlation

Answer 14

Non-parametric test. Can be used with ordinal data, or continuous data that are not normally distributed

Type: data must be ordinal, interval or ratio level of measurement
Ties: participants variable levels should not be the same across multiple people

Question 15

Kendall’s Tau rank correlation- tau

Answer 15

Often used instead of spearman’s when there are tied scores
Types: data must be ordinal, interval or ratio level of measurement

Question 16

Covariance

Answer 16

Covariance measures this: the extent to which variables co-vary
High covariance means there is large overlap between the patterns of change observed in each variable
Low covariance means there is little overlap in the variance of each variable