Research methods and statistics in psychology Flashcards
Primary purpose of psychological research?
test hypotheses (i.e., statements of cause-and-effect that are derived from a given theory and tested in research)
Hence is theory based
What are some research methods that can be used?
- Experimental Method
- The quasi-experimental method
- The survey (or correlational) method
- The case study method
Describe the experimental method
common research method that manipulates a theoretically relevant variable and to examine the effect of this manipulation on an outcome variable
manipulation a = outcome b
What are the different components of the experimental method?
Experimental Group - the group receiving treatment
Control Group - compared to
Experimental treatment - the administration of experimental treatment in experimental group is a difference between experimental and control gorpu
Random allocation of subjects are made in accordance to the conditions (condition that experimental group is in and condition that control group is in)
Why is there random allocation of subjects to experimental and control group?
Lower chance of systematic differences such as intelligence, personality
- Better representation of population which means the theory can be better generalized
- unsystematic allocation of participations to conditions
What is a method that may be used for random allocation?
What may not be a method to use
Flipping a coin
First 50 followed by next 50, assigning numbers
Experimental Method
What is the difference between the dependent and independent variable?
Dependent variable is fundamentally, the outcome of the experiment that relies on the treatment
Independent variable is the manipulation introduced by the researcher and is known as the **treatment variable **
E.g., 1: Does my physical training regime increase cognitive processing? ?
IV = physical training, DV = cognitive processing
, 3: Does watching violent TV result in more violent behaviour?
DV = degree of violent behaviour, IV = amount of violent TV
What are the strengths and weaknesses of the experimental method?
Strengths:
- Researchers have control of IV in experiments
- when there is a significant change in the DV, Causal inference can be made has to be a result of the manipulation of IV
- by appropriate use of statistics and sufficiently large experimental samples, results can reflect the larger, general population. Participants representative of general population
Weaknesses:
- Results from experiments can be invalid when IVs don’t really manipulate what researchers think they’re manipulating.
Threat to internal validity that may be through experimental confounding (unintended manipulation of IV)
such as heat being a confound in the experiment to measure whether crowding causes stress, diet, age, the participant’s knowledge they are participating in an experiment.
When is the quasi-experimental method used?
When random allocation of subjects in groups cannot be achieved because it’s impossible or for unethical reasons
What is the quasi-experimental method and how does it differ from the experimental method?
It does not use random allocation of subjects, if so, the casual inference will be invalid
This is because the independent variable cannot be manipulated eg. gender, age
What are variables that cannot be controlled in a study?
Extraneous Variables
Confounds are part of extraneous variables
Strengths and weaknesses of quasi- experimental method?
Compared to the experimental method, there is much less control over the IV (and hence we can’t conclude that it is necessarily responsible for change in the DV).
Despite the fact we try and Match our groups as closely as possible
What research method may be used when finding naturally occurring patterns of variables in real world? give an example
survey or correlational method
wealth and attitudes to education
Strengths and weaknesses of the survey method
- Make predictions based off results
- Discover issues in real world
- explores number of different variables and their correlation
- useful for making predictions, based on observed relationships (e.g., that people who are overly stressed will suffer heart attacks).
Weakness:
- Results are generally more descriptive than those obtained using experimental methodology.
- Does not provide evidence to results as an experimental method would
- Correlation does not equate to causation
What research method measures individuals? Give an example
Case studies
Brain trauma
Advantages of case studies?
Very detailed providing insight to an individual’s mental processes and behaviour
Why is choice of different methods beneficial?
In view of the potential limitations of any one method, many researchers use (or at least consider using) multiple research methods in order to explore the same issue in many different ways
How can experimental pitfalls be overcome?
good experimental design such as counterbalancing and using appropriate control groups
Distinguish Between subjects vs Within Subjects designs
between subjects: The experimental manipulation occurs between the control and experimental groups
within subjects:
The same subjects are exposed to the IV at different times
i.e., the experimental manipulation occurs within the subjects.
Reliability in psychological research
Reliability: same or similar results will be produced when the study is conducted again in the future
Validity in psychology
the study measures what it is meant to measure
What are three types of validity
Internal Validity: does the outcome really reflect the experimental manipulation?
- What was the experimental manipulation, where and how was the study conducted
External validity: How well can we generalise?
- .Who was the subject population
Construct validity: Correspondence between the theory (construct) and the measurement used.
- . Was only one measure* used or were *multiple measures used to achieve triangulation
What is goal in selecting IV and DV?
in selecting an IV and DV to identify that variable which approximates most closely to the theoretical variable of interest.
eg. unethical to induce depression, use melancholy music instead to achieve desired DV
How can improper selection of DV and IV influence validity
Improper choice or manipulation of the IV and/or DV can result in serious violations of internal and external validity.
What can threats to internal validity lead to?
researchers interpret effects they obtained incorrectly
Threats to internal validity
What does maturation effects refer to?
when we use a situation where time passes,
The longer the time between the pre-test and post-test, the more likely the post-test is likely to reflect maturation effects
Threats to internal validity
History effects
changes in society
How might the testing situation itself result in threats to interval validity?
Testing effects reflect the situation where for one reason or another, the original testing situation (O1) influences the subsequent testing situation (O2 ).
Participants might be less motivated in the second half “here we go again”, or they might be more motivated “I want to do better this time” (psychological changes)
The therapist’s room may have been re-decorated (and made less drab). (physical changes)
Threats to internal validity
Hawthorne effect
subjects simply respond to he fact that they are part of the research
Threats to internal validity
mortality effects
subjects drop out of study
What are experimental designs that can overcome internal validity threats?
involves getting rid of the first observation, so that there is no pre-test for the experimental treatment to react with:
This control condition involves different participants but a pre-test and post-test that are identical in every respect:
What is experimenter bias and How can this bias threaten external validity?
Any behavior by the researcher that prevents a fair test of the experimental hypothesis.
The experimenter may react more positively to responses that are to his or her liking (for example, by nodding or smiling)
- harder for experimental results to be reflective of external world
How can problems with demand characteristics be overcome?
–Problems associated with demand characteristics can also be addressed by concealing the purpose of an experiment. This can be done by means of deception — either misinforming participants about the purpose of an experiment or using concealment to avoid telling them the whole truth.
Threats to internal valdity
Fatigue Effects
even short delays between pre and post-test
eg. children get tired quickly
Threats to external validity
demand characteristics
demand characteristics: subjects react to cues in research which seem to indicate how they ought to respond.
Person may respond to the question ‘Are you sure you’re anxious?’, by saying ‘No’ (when they actually are anxious), because they think the response ‘No’ is being demanded of them.
What is difference between survey and experiment?
in an experiment we manipulate the IV, and measure the DV.
In a survey, we measure **what we think may be causing the behaviour **(the IV) and also measure the behaviour (the DV).
what characteristics do surveys exhibit?
- naturally-occurring and enduring relationships between variables
- description than with explanation.
- representative of population of interest
Surveys:
What does correlation does not equate to causation suggest?
no relationship between variables this allows us to conclude that one does not cause the other
Survey
What are design issues survey may encounter?
a) sample selection
b) sample size
c) Type of surveys
Surveys
How can sample selection be a design issue?
To generalize results to the larger population, survey researchers need to ascertain their sample is representative random sample of population.
–survey must obtain data from a random sample of the population that has the same characteristics as the population.
Example of when sample selection lead to wrong prediction
–the Literary Digest
Survey Design Issues
Sample Selection: What does convenience sampling refer to?
–generally the population in which researchers are interested is all people who display the psychological process they are investigating.
No matter how narrowly people are chosen, they should be representative of this population
Survey
Design Issues: Sample Size
Benefits and disadvantages
–larger sample size reduces uncertainty about the inferences drawn from sample data. More reflective of population
Disadvantages of larger sample size
- systematic bias can occur
–size will be reduced by non-responses.
- mortality and reactivity. If particular people (e.g., lazy ones, or people who are offended by the survey) don’t respond this can bias the sample.
Survey
Design Issues:Type of surveys
Limitations
- Socially desirable responding: people respond in ways they think they should, rather than as they normally would .
- In this way, data can be biased by a concern to provide socially desirable responses.
- Reactivity
- Mortality rate
What is a way to overcome problem of demand characteristics in surveys?
use non-obtrusive measures, which ensure that the participants are not aware that they are being involved in the research process
Overall, what are survey designs that can help overcome limitations?
ncarefully worded so as not to be misleading.
what is triangulation? what is the benefit?
Triangulation in research increases the credibility of the research by drawing on multiple viewpoints.
**Increases construct validity of research **
construct validity
Construct Validity refers to the ability of a measurement tool (e.g., a survey, test, etc) to actually measure the psychological concept being studied.