Note
Studied by 21 people
Ch 2
THE HYPOTHESIS TESTING APPROACH
THEORIES AND HYPOTHESES
Most personality research begins with a theory
A general statement about the relationship between constructs or events
Some psychologists use theories that are very broad in application
e.g., using Freud's psychoanalytic theory to explain what causes psychological disorders, why people turn to religion, or to why people find certain jokes funny
Personality researchers work with theories narrower in application
e.g., speculating about the reasons why some people are more motivated to achieve than others or the relationship between a parent's behaviour and a child's level of self esteem
Broad theories, like Freuds, can be seen as collections of more specific theories that share certain assumptions about the nature of human personality
Characteristics of a good theory:
A good theory is parsimonious (parsimony)
The simplest theory that can explain the phenomenon is the best
If two theories can account for an effect equally well, the simpler explanation is preferred
A good theory is useful
Unless a theory can generate testable hypotheses, it will be of little or no use to scientists
Although, ideas that cannot be tested are not necessarily incorrect
They simply do not lend themselves to scientific investigation
Unless a theory is testable, it cannot be examined through scientific methods
Thus, holds little value to scientists
Theories themselves are never tested
Hypotheses are derived from theories that can then be tested in research
Hypothesis: is a formal prediction about the relationship between two or more variables that is logically derived from a theory
Since theories are never tested directly, they are never proved or disproved
Thus, theories are more or less supported by research and therefore is more or less useful to scientists trying to understand the phenomenon
More often research confirms predictions = higher confidence in how accurate the theory describes the nature of things
Empirical investigation continuously fails to confirm predictions = less likely we are to accept the theory
With this, scientists modify the old theory or generate a new one
EXPERIMENTAL VARIABLES
Independent variable: determines how the groups in the experiment are divided
This is manipulated by the experimenter
Generally uses different levels of the IV to create the experimental groups
I.e., the treatment variable
Dependent variable: measured by the investigator and used to compare the experimental groups
Differences between groups on the DV can be attributed to the different levels of the IV
I.e., the outcome variable because researchers want to say that the differences in the DV are the result of the changes made in the IV
Researchers typically use multiple IVs
A result of using multiple factors could be an interaction
How one IV affects the DV Is dependent on IV
e.g., whether anxiety (IV) leads to an increase or decrease in questions (DV) is dependent on when the participant is high or low in shyness (IV)
MANIPULATED vs NONMANIPULATED INDEPENDENT VARIABLE
A researcher who uses a manipulated independent variable begins with a large number of participants and randomly assigns them to experimental groups
Each person has an equally likely chance of being assigned to group A as they do B
Individual differences are accounted for with the law of large numbers
By using a large number of participants and randomly assigning them to conditions, researchers assume that the differences will even out
Now the IV is introduced
Some watch violent T.V, others watch nonviolent programs
Because it is assumed participants in each condition are nearly identical on average at the start of the study, any differences in the groups AFTER watching the program can be attributed to the IV
A nonmanipulated independent variable (I.e., subject/participant variable) exists without the researchers intervention.
e.g., splitting a group into first-born, second-born, and third-born children
With this, the investigator does not randomly assign participants to a condition
Each participant already belonged to one of the groups, and the researchers simply determined which group that was
The problem with nonmanipulated IVs?
The researcher cannot assume the groups are nearly identical at the beginning of the experiment (non-equivalent groups)
Therefore, difference could be attributed to individual differences and not T.V programs
Manipulated variables are preferred but not always possible
This is a particular problem in personality research, as many variables cannot be manipulated
PREDICTION vs HINDSIGHT
We expect researchers to make predictions about that will happen in study before the data are collected
A hypothesis cannot be generated AFTER the results were seen
E.g., who is more impressive? Someone who can explain after a basketball game why the winning team was victorious? Or the one who tells you before the game which team will win, and why?
The latter; anyone can come up with an explanation after the fact, but people who really understand the game can make reasonable guesses about what will happen when the 2 teams meet
REPLICATION
It is dangerous to assume that a significant finding from one study provides reliable evidence for an effect
This problem is dealt through replication
The more often an effect is found in research, the more confidence we have that it reflects a genuine relationship
BUT determining the strength of an effect by how often it is replicated is not always easy
Well-known research findings may be difficult to replicate
But because the failures at replication are stored away in file drawers, we might not realize the problem exists
THE CASE STUDY METHOD
An in depth evaluation of a single individual (or sometimes a few)
Typically, the participant is suffering from a problem of interest to the investigator
The persons history, current behaviour, and changes in behaviour over the course of investigation
Case study data is usually descriptive
Rather than reporting a lot of numbers and statistical analyses, investigators describe their impressions of what the person did and what it means
Occasionally include quantitative assessments
Case studies have played an important role in the history of personality psychology
Weaknesses:
It is difficult to generalize from a single individual to other people
Just because one person reacts to events a certain way, does not mean other people do
It is difficult to determine a cause-and-effect relationship with the case study method
The accuracy of some case study observations is questionable
Researchers form expectations --> causes them to see what confirms their hypotheses --> overlook disconfirming evidence
Despite the weaknesses, case studies are used because:
Other research methods might not do the job
Quantitative methods may not capture the richness of a person's life
Valuable for generating hypotheses about the nature of human personality
Useful for rare cases
Appropriate when researchers can argue that the individual being studied is essentially no different from all normal participants on the dimension of interest
Illustrate a treatment
Demonstrate possibilities (I.e., high ends of the spectrum/exceptions)
STATISTICAL ANALYSIS OF DATA
STATISTICAL SIGNIFICANCE
More common tests include analysis of variance, a chi-square test, and a correlation coefficient
If scores differ by an amount so small that it could have been caused by a chance fluctuation, the difference has not reached statistical significance
If the difference is so large that in all likelihood, it was not caused by chance but reflects a true difference, the difference is statistically significant
Psychologists use a significance level of 0.05 to assess
If the difference between the scores is so large that it would occur less than 5% of the time by chance, the difference is probably genuine
CORRELATION COEFFICIANTS
Can range from +1 to –1
PERSONALITY ASSESSMENT
Valid and reliable assessment tools are important
e.g., when studying achievement motivation, self-esteem, social anxiety, they must as accurate as possible
It is the responsibility of those using the test to see that it accurately measures the concept of interest
RELIABILITY
A test has good reliability when it measures consistently
Measures something consistently, but doesn't tell us WHAT is being measured consistently
A reliable personality test will produce roughly the same score each time you take it
Inconsistent scores may be due to:
Scoring procedure may be vague
How test takers take the test depends on mood
But since we assume personality is relatively consistent over time, tests designed to measure personality should provide consistent scores
Test-retest reliability coefficient:
Large group takes a test --> weeks later, they take the test again --> correlation coefficient is calculated between the 2 scores
A high correlation coefficient indicates good consistency over time
Whether or not a moderate correlation is considered "high enough" depends on the concept and researchers needs
Internal consistency
A test is internally consistent when all items on the test measure the same thing
VALIDITY
Refers to the extent to which a test measures what it is designed to measure
Hypothetical constructs: useful inventions researchers employ to describe concepts that have no physical reality
The issue lies in measuring something that is abstract/hypothetical
Face validity
What a test appears to be measuring
Good or poor face validity
e.g., creativity vs. Social anxiety
Congruent validity (I.e., convergent validity)
The extent to which scores from the test correlate with other measures of the same construct
Discriminant validity
Refers to the extent to which a test score DOES NOT correlate with scores of theoretically unrelated measures
e.g., does the test actually measure creativity? Or another construct that simply resembles creativity, like intelligence?
Give two tests, one measuring creativity, the other intelligence --> high correlation between the scores --> does not measure what is intended
Behavioural validation
The test scores predict relevant behaviour
e.g., it is possible test takers respond to assertiveness scales by indicating how they think they would act or how they wish they would act, rather than how they would actually act
It is possible to demonstrate a tests face validity, congruent validity, and discriminant validity, yet still not have a valid measure if unable to predict how people would act in relevant situations.
