Unit 1- Research Methods Flashcards
Human intuition is highly
Limited
Overconfidence
we tend to think we know more than we do
Hindsight Bias
Tendency to believe, after learning an outcome, that one would have foreseen.
The “I know it all along” phenomenon
Theory
An explanation unhinge an integrated set of principles that organizes and predicts observations
Research methods: Case study
An observation technique in which ONE person is studied in depth in the hope of revealing universal principles.
Freud relied on it
Research methods: survey
Technique for ascertaining the self reported attitudes or behaviors of people
Usually by questioning a representative, random sample of people
Random sample
Everyone in the population has an equal chance of being selected
Anchoring bias
We tend to rely too heavily on the initial information (anchor)
Advantages of case study
In depth, detail of person being studied
Opportunity to study unusual cases
Disadvantages of case study
Individual results might not relate to general public
Cannot establish cause and effect
Components of survey: population
All individuals you are interested in knowing something about
Components of survey: sample
Individuals you actually question. People in survey
Components of survey: why should sampling always be taken randomly from the population?
So that it is representative meaning each individual in the population had an equal chance of being selected
Advantages of survey research
Less costly to run that an experiment
Efficient, allow researchers to collect lots of information quickly
Disadvantages of survey research
Accuracy can be thrown off by improper sampling
Question wording can impact results of a survey
Respond rates may be low
Participants can lie
Stratified sampling
When the researcher divides the sample into subgroups called strata.
Research method 3: naturalistic observation
Observing and recording behavior in naturally occurring situations without trying to manipulate and control the situation
Drawbacks: hard to identify any type of causation since there is no control.
Research method 4: Correlation research
Research that looks at a RELATIONSHIP between two variables (things)
Types of correlations
Positive and negative correlation
Positive correlation
1 goes up the other one goes up
1 goes down the other one goes down
Negative correlation
One goes up the other one goes down
One goes down the other one goes up
Correlation coefficient (r-value)
Number that measures the strength of correlation
Strongest correlations are ____ and ____
+1 and -1
Correlations are always between -1 and +1
A graphed cluster of dots, each of which represents the values of 2 variables
Little _____ indicated high correlation
Scatter plot
Illusory correlation
The perception of a relationship where none exists.
Advantages of correlation research
- Allows you to see a relationship
- efficient, don’t have to complete whole new study(use existing data)
- provide way to make predictions
- help dispel illusionary correlations
Disadvantages of correlation research
- do not show causation
- prone to inaccurate reporting
- other variables may have impact on relationship
Summing up the researches
All of these methods look to describe the behavior not to explain it
None establish causation
Experimental designed research
Only research that gets out causation
Why do we make a thinking error by perceiving order in random events
Random sequence don’t often look random
Research method 5: the experiment
An investigator manipulates 1 or more factors (IV) to observe their effect on some behavior or mental process (DV) while having other factors by random selection
Operational definition
A statement of operations used to define research variables
Components of an experiment
IV-experimental factor and DV-behavior or mental process
Steps for a perfect experiment
After variables,
1) find participants by random sampling from a representative population
2) randomly assign the participants to the control and experimental conditions
Groups within an experiment: experimental condition
The group of participants that is exposed to the independent variable (treatment)
Groups within an experiment: control condition
Group of participants that does not receive the independent variable (treatment vs no treatment)
control condition placebo
Inert substance (usually in form of a fake pill)
Steps for a perfect experiment continued
- Set up control procedures to stop possible bias and confounding variables
Steps for a perfect experiment continued: double blind procedure
Both the research participants and the research staff are blind (ignorant) whether the research participants have received the treatment or a placebo.
Being bias
Trying to purposely prove your results
Experiments want to avoid: confounding variables
A hidden or uncontrolled aspect of an experiment that can resort the results of an experiment
Clever Hans
Shows importance of control procedures like double blind procedures
Last step of a perfect experiment
4) compared the results using statistics between the experimental condition and the control condition and see if the differences between the 2 groups of the experiment are statistically significant
statistically significant
Experimental results are not likely to have happened by random chance;helps establish causation
Usually expressed as a number called a p-value
5th step
Investigators should replicate the experiment again with different participants
Research strategies: replication
Repeating the essence of a research study to see whether the basic finding generalized to other subjects and circumstances
When experiment goes wrong:
Every time you are conducting an experiment, you are trying to prove or reject your hypothesis
Type 1 error: false-positive
Results show that difference is not true
Pregnancy test says a woman is pregnant when she is not a type 1 error
Type 2 error: false negative
Results fail to detect difference when there is one
Pregnancy test says a woman is not pregnant when she is shows a type 2
Types of statistics psychologists use: inferential statistics
Making generalization beyond the general data to try to say something about a broader population.
Stating the American’s public approval of the president after surveying 5000 people involves inferential statistics
Types of statistics psychologists use: descriptive statistics
Information about the data you have already gathered
descriptive statistics: measures of central tendency: mode
most frequently occurring
Mean: average of a distribution
average of a distribution
Median
The middle scores in the distribution
When the mean is distorted
Inferences aren’t true
Positive skew and negative skew
Positive skew- mean is bigger than median
Negative skew- median is smaller than mean
Descriptive of statistics continued: range
The gap between the highest and lowest scores.
Shows extreme scores
Standard deviation
How much scores vary around the mean.
Shows how packed together the scores are around the average
Ethical procedures in research
1) participants should not be out at risk psychologically or physically
2) participants have the right to privacy
3) deception is allowed in experiments but are scrutinized by ethics boards (institutional review board)to decrease risks to subjects.
4) researchers must debrief participants about the full nature of the research immediately after participation.
5 and 6 procedures
- Informed consent
6. Permitted to leave at one point
Advantages of experimental research
- Establishes cause and effect
- Allows IV and DV to be closely controlled and manipulated
- Allow double-blind method to control bias
- Can be replicated
Disadvantages of experimental research
1) Experimental environment may be artificial and not relate to real world.
2) may be difficult to eliminate all confounding variables
- Hawthorne effect
People’s behavior tends to change just by the fact that they’re being watched
Sampling bias
Any non random technique
