# Psychology Correlation

• Created by: izx.a
• Created on: 18-02-18 12:39
Positive Correlation
As one co-variable increases, the other co-variable also increases
1 of 40
Negative Correlation
As one co-variable increases, the other co-variable decreases
2 of 40
Zero Correlation
There is no correlation between the co-variables
3 of 40
Correlation co-efficients
Shows the type of correlation. The measure of the strength of correlation between -1 and +1
4 of 40
Do correlations establish cause and effect?
No. Correlations only establish relationships between two co-variables
5 of 40
Co-variable
Two variables you're looking at for a relationship
6 of 40
1 tailed correlational hypothesis for ice cream sales and hours of sunshine
There will be a significant positive correlation between ice cream sales per week and hours of sunshine peer week.
7 of 40
Null correlational hypothesis on time spent using phone and spelling ability
There will be no significant relationship between time using phone per week (in hours) and someone's spelling ability meansured using their score on a spelling test. Any correlation will be due to chance.
8 of 40
Are correlations quantitative or qualitative data?
Quantitative. They produce numerical data.
9 of 40
Which two levels of data do correlations deal with?
Ordinal and Interval/ratio
10 of 40
Are correlations expressed as bar graphs or scatter graphs and why?
Scatter graphs. This makes it able to plot individual results and to establish a relationship between the results.
11 of 40
What's the difference between correlation and an experiment?
Correlations don't establish cause and effect and only look at a relationship between two co-variables. Experiments predict DIFFERENCE in results from participants in different conditions.
12 of 40
Research question and example
A broad question about the concept being investigated (eg. Are people happier when it's sunny?) Must end with a '?'
13 of 40
Research aims
Identifys a more specific concept than the research q. Can take many forms. The aim will have an impact on the research method that the researcher goes onto use (correlational, experimental)
14 of 40
Examples of research aims
1. To investigate whether there is a relationship between sunshine and happiness. 2. To investigate whether sunshine is a cause of happiness.
15 of 40
Operationalising variables
Making the variable measureable/testable. Any co-variable needs to be expressed as a number. Each participant must have two separate (ordinal level) numbers that relate to them - one for each co - variable.
16 of 40
Two tailed correlational hypothesis
"Significant correlation between co-variables y and z". Significant relationship predicted but not including expectational words (positive or negative). No direction is given.
17 of 40
A one tailed correlational hypothesis
Indicate researcher's direction expectation. "Significant positive correlation between co variables y and z"
18 of 40
Null hypothesis
"There will be no significant correlation between co-variables y and z; any relationship will be due to chance factors"
19 of 40
Primary data
Data gathered directly from the participants by the researcher
20 of 40
Secondary data
Data that has already been gathered by someone other than the researcher
21 of 40
Which type of data is used in correlational studies?
Researchers could be using either forms of data, unlike when carrying out experiments or self-reports where primary data is mostly used.
22 of 40
Sampling method
Researcher will need to decide what the target population is that they want to generalse to and settle on a way of obtaining a sample of either participants or data that will be representative of this
23 of 40
Ethics
If collecting primary data, researcher will need P's consent, avoid decieving them, give them chance to withdraw their data etc.
24 of 40
ethics - secondary
If using secondary data, it may already be that it is in the public domain. If so, the researcher will still need to ensure that the people whose data is being used are not identifiable in any way and that these results cannot harm them in any way.
25 of 40
Interpreting scattergraphs
A scattergraph would never justify you in concluding that co variable X has had an effect on co variable Y, only their relationship between each other.
26 of 40
Interpreting scattergraphs 2
A scattergraph would never justify you in saying that a particular hypothesis can be retained or rejected
27 of 40
Interpreting scattergraphs 3
Don't confuse findings with conclusions
28 of 40
Findings
relate to raw data (so you could point out mode, median or mean, the range of responses for a co - variable, or point to a particular outlier).
29 of 40
Conclusions
relate to broad inferences that you can make from that raw data (eg positive, negative or no correlation - and also the approx strength of any correlation).
30 of 40
Inferential statistics for correlations
To enable a researcher to find out whether they have a correlation or not (and, therefore, to know which hypothesis to reatin or reject) they use these tests (eg Spearman's Rho test).
31 of 40
Inferential statistics tests
These enable researchers to calculate a 'correlation co-efficient' that will be somewhere beteen -1 and +1.
32 of 40
What does this mean?
A correlation that has a plus sign (+0.58) is a posi correlation. One with a minus sign (-0.72) is a negi correlation. One with a co-efficient around 0 (eg +0.12 or -0.17) represents no correlation.
33 of 40
Interepreting data from correlation study
Main thing to remember is that the further from 0 the correlation co-efficient is (in either a posi or negi direction) the stronger the correlation is.
34 of 40
Strength of correlation
How strong the effect is iwll determine which hypothesis the researcher holds onto - either alternative or null hypothesis.
35 of 40
Although they don't involve collection of new data, they tell us something new; new relationships learned about pre existing data. Can also tell us the direction of a relationship and the strength of it, too.
36 of 40
No manipulation meansw its useful when either practical or ethical reasons means that variables cannot be manipulated.
37 of 40
Can act as a good starting point for research because, once relationships have been established, more research (eg experiments) can be conducted to investigate them further.
38 of 40
Do not tell us anything about cause an effect. Cannot tell us what is causing the effects show between the two co variables. May be hidden factors casuing the relationship.
39 of 40
The inferential statistics tests will not always pick up on a relationship between co-variables.
40 of 40

## Other cards in this set

### Card 2

#### Front

As one co-variable increases, the other co-variable decreases

#### Back

Negative Correlation

### Card 3

#### Front

There is no correlation between the co-variables

### Card 4

#### Front

Shows the type of correlation. The measure of the strength of correlation between -1 and +1

### Card 5

#### Front

No. Correlations only establish relationships between two co-variables