Correlational Research Correlational Research Inferential/Descriptive Statistic (r) Describes strength of linear relation between two variables Strength of relation = degree of association Does knowing an individual’s score on A tell you much about that individual’s score on B? How much does it tell you? Some properties of r Ranges between -1 and 1 inclusive Sign (+ or -) gives direction of relation Absolute value gives strength r 2 gives proportion of variance accounted for
28
Embed
Correlational Research Inferential/Descriptive Statistic (r) Describes strength of linear relation between two variables Strength of relation = degree.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Correlational ResearchCorrelational ResearchInferential/Descriptive Statistic (r) Describes strength of linear relation between two
variablesStrength of relation = degree of associationDoes knowing an individual’s score on A tell you much about that individual’s score on B? How much does it tell you?
Some properties of r
Ranges between -1 and 1 inclusive
Sign (+ or -) gives direction of relation
Absolute value gives strength
r2 gives proportion of variance accounted for
CorrelationsCorrelations
Performance IQ
14013012011010090807060
Ve
rba
l IQ
140
130
120
110
100
90
80
70
60
Correlation between VIQ and PIQ = +.53
Characterized as strong and positive
CorrelationsCorrelations
Correlation between C and Behavioral Impulsivity = -.41
Characterized as moderate-strong and negative
underco, i/i, mr, teaimp, eysenck
86420-2-4-6
con
sice
ntio
usn
ess
8
7
6
5
4
3
2
CorrelationsCorrelations
Correlation between C and Family SES = -.15
Characterized as weak-negligible
family social class
6050403020100
con
sice
ntio
usn
ess
8
7
6
5
4
3
2
Correlational ResearchCorrelational Research
Disadvantages:1) correlation does not indicate causation2) problems with self-report method
Advantages:1) can collect much information from many subjects at one time2) can study a wide range of variables and their interrelations3) study variables that are not easily produced in the laboratory
Seeks to establish a relation/association/correlation between two or more variables that do not readily lend themselves to experimental manipulation
Disadvantages:1) correlation does not indicate causation (6slide
causation example)2) problems with self-report method Advantages:1) can collect much information from many subjects at one time2) can study a wide range of variables and their interrelations3) study variables that are not easily produced in the laboratory
Microsoft PowerPoint Presentation
Correlation is not CausationCorrelation is not CausationThree ways in which A and B may be causally related
Specify and measure likely third variablesControl for third variables in model
If association between A and B remains, then third variable explanation does not hold
If A and B are causally related, they must be correlated
Can model third variable effects
Longitudinal design
Experimental design
Appeal to strong theory
Case In PointCase In Point Lynam, D.R., Moffitt, T.E., & Stouthamer-Loeber, M.A. (1993).
Explaining the relation between IQ and delinquency: Class, race, test motivation, school achievement, or self-control? Journal of Abnormal Psychology, 102, 187-196. IQ is negatively correlated with delinquency
Think of some ways this relationship could work. In other words, why or how might lower IQ be related to delinquency?
What are other possibilities?
Delinquency leads to low Delinquency leads to low IQ?IQ?
Third Variables?Third Variables?
What are the likely candidates?
Social Class
Impulsivity
Race
What happens when they are included?
Partial r = -.15
Original r = -.25
Significant relation remains
IQ IQ Delinquency via School Delinquency via School Achievement?Achievement?
Seeks to establish a relation/association/correlation between two or more variables that do not readily lend themselves to experimental manipulation
Disadvantages:1) correlation does not indicate causation2) problems with self-report method
Advantages:1) can collect much information from many subjects at one time2) can study a wide range of variables and their interrelations3) study variables that are not easily produced in the laboratory
Microsoft PowerPoint Presentation
Microsoft PowerPoint Presentation
Correlational Research Correlational Research ExampleExample
Reformulated learned helplessness:
When people find they are helpless, they ask whythree dimensions of causal attributions
internal-externalspecific-globalstable-unstable
internal, global, and stable attributions lead to depression
You have been looking unsuccessfully for a job for some time.1. Write down one major cause.
2. Is the cause of your unsuccessful job search due to something about you, or to, something about other people or circumstances ?Totally due Totally due to others 1 2 3 4 5 6 7 to me
3. In the future, when looking for a job, will this cause again be present? Will never again Will alwaysbe present1 2 3 4 5 6 7be present
4. Is this cause something that influences just looking for a job, or does it also influence other areas of your life? Influences this Influences allsituation 1 2 3 4 5 6 7situations
5. How important would this situation be if it happened to you? Not at all Extremelyimportant1 2 3 4 5 6 7important
Definition: Blends correlational and experimental approaches. Typically examines interaction between an individual difference variable and an experimental manipulation(expericorr example)
Disadvantages: similar to those of correlational research
1 Third variable problem still remainsAdvantages: similar to those of correlational research
1 Can tease out cause and effect somewhat better if have manipulated variable
Definition: Blends correlational and experimental approaches. Typically examines interaction between an individual difference variable and an experimental manipulation
Disadvantages: similar to those of correlational research
1 Third variable problem still remainsAdvantages: similar to those of correlational research
1 Can tease out cause and effect somewhat better if have manipulated variable
Locus of Control ScaleLocus of Control ScaleFor each pair of statements, choose the one with which you most agree:
1a. Many of the unhappy things in people's lives are due partly to bad luck.1b. People's misfortunes result from the mistakes they make.
2a. One of the major reasons we have wars is that people don't take enough interest in politics.2b. There will always be wars, no matter how hard people try to prevent them.
3a. Sometimes I can't understand how teachers arrive at the grades they give.3b. There is a direct connection between how hard I study and the grades I get.
4a. The average citizen can have an influence in government decisions.4b. This world is run by the few people in power and there isn't much the little guy can do about it.
Scoring: Give yourself one point for each of the following: 1a, 2b, 3a, 4b
Locus of ControlBeliefs about one’s ability to influence outcomesSimilar to learned helplessnessAssessed with LOC scale
Selected participants to be high or low in external LOCRandomly assigned participants to one of two conditions
Performance task with shock that is very easy to learn to avoidPerformance task with shock that is hard to learn to avoid
Findings in line with hypothesesIn easy task most people learned how to avoid shockIn harder task, Ps with high external LOC were less likely to learn
Microsoft PowerPoint Presentation
Problems to Look ForProblems to Look ForExperimental confound: when another variable is allowed to vary with the independent variable.
Third variable problem
Nonrandom samples: when samples are selected so that not everyone has an equal opportunity to participate.Failure to replicate: single experiments do not constitute facts.
Failure to include relevant comparison groups: lack an adequate baseline or normative data.
(Example)
Failure to Include Failure to Include ComparisonComparison
Make people happy or sad and examine effects on aggression: