Research Methods Flashcards
Hypothesis
A hypothesis is a testable prediction that is falsifiable.
Three categories of research methods?
descriptive, correlational, and experimental research.
descriptive research
The goal of descriptive research is to simply describe or measure the variables that you are interested in (e.g., the level of gratitude and happiness in our PSYC 130 class).
Variable
any characteristic (e.g., eye color, personality trait, willingness to help in an emergency) or condition (e.g., time of day, temperature) that can change or take on different values.
Three types of descriptive Research
- Observational research.
- Case studies.
- Surveys.
Naturalistic observational studies
- Observations take place in a natural setting (i.e., a setting that has not been structured by the researchers).
- Allows you to observe behavior as it occurs naturally.
- Time consuming. Researchers must wait for the behavior they are interested in to occur naturally.
Structured observational studies
The setting is structured by the researchers with the intention of eliciting the behavior of interest. - Structured observational studies can take place in the lab or in real-world settings.
- Major advantage of not having to wait for the behavior of interest to occur naturally.
- Disadvantage is that thebehavior of interest may be less natural because the environment has been structured by the researchers
Participant observational research
- The researchers actively engage and interact with the participants they are observing.
- Allows for observations of behavior that may be impossible to observe otherwise.
- Potentially dangerous and time consuming. - Researcher involvement may influence the behavior of the participants, so it is less natural.
Case Studies
In-depth examination of an individual or small group. The goal is acquiring detailed information.
There are many different tools for collecting information in a case study (e.g., observations, psychological tests, surveys, interviews, selfreport measures).
Useful for studying rare cases that do not occur often and are impossible or unethical to manipulate (e.g., H. M’s removal of his hippocampus).
Findings from case studies can be used to generate hypotheses that can then be tested using more people and other research methods.
- Can be used to provide evidence that a psychological phenomenon is possible (i.e., existence proof).
Weakness of Case Studies
Weakness of case studies are they are not generalizable. The information uncovered in a case study may not generalize to other people or other situations. In other words, case studies are anecdotal.
Surveys and Self-Report Measures
Goal is to describe the attitudes, experiences, beliefs, and behaviors of a population of people.
Allow researchers to describe and measure variables that would be impossible or unethical to observe directly (e.g., frequency of sexual harassment experienced by college students).
Can be an easy and quick way to gather data from many people.
Concerns of surveys
Reliance on selfreported data (i.e., data that people report themselves). Assumption is that people are being honest and accurate in their responses. - Social desirability bias (i.e., people may want to be viewed in a positive manner and therefore, adjust their responses to appear more desirable).
Self-report psychological measures
are related to surveys because they both rely on self-reported data. A selfreport measure is designed to measure participants on a psychological concept by asking them to respond to questions that are designed to capture that concept (e.g., NPI-40; a self-report measure of narcissism).
Correlational Research
The goal of correlational research is to examine and measure the relationships (i.e., association) between variables.
Correlational research involves measuring variables and the association between them (e.g., the relationship between gratitude and happiness scores).
the Pearson Product Correlation Coefficient (r).
The extent to which the scores on two variables are associated can be measured statistically
A correlation coefficient
provides us with the statistical strength and direction of the relationship between two variables.
- Correlation coefficients can range from -1 to +1.
- A positive correlation coefficient
indicates that as the scores on one variable changes, the scores on the other variable tend to change in the same direction (e.g., as gratitude scores increase, happiness scores tend to increase as well).
negative correlation coefficient
indicates that as the scores on one variable changes, the scores on the other variable tend to change in the opposite direction (e.g., as gratitude scores increase, narcissism scores tend to decrease).
correlation does not equal causation.
Remember
Experimental Research
Experimental research allows us to determine if there is a causal relationship between variables
- The manipulation of an Independent Variable (IV)
The researchers vary the IV across different conditions of the experiment.
- The measurement of a Dependent Variable (DV).
The researchers measure the DV across the different conditions of the experiment to determine if changes in the IV caused changes in the DV.
- Random Assignment
Every participant has an equal chance of being assigned to the different conditions of the experiment. Random assignment ensures that the pre-existing differences of the participants are spread out evenly across the conditions of the experiment.
- Control of potential confounding variables.
Confounds are variables, other than the IV, that differ across the conditions of the experiment, and have the potential to influence the DV. Researchers must eliminate confounds to ensure that any differences found in the DV across the conditions of the experiment can be attributed to the manipulation of the IV and not something else.
Selecting Participants
The first step in research
your population of interest
the people that you are interested in learning about).
From your population of interest, you will select a
sample
Random Selection
(i.e., every person in your population of interest has an equal chance of being selected for your sample).
uses Convenience
Sampling
(i.e., a sample is simply made up of people who are available and willing to participate).
Operational definitions
are the precise ways that conceptual variables are measured or manipulated in a study.
Construct Validity
is the extent to which the measures used in a study capture the variables they were designed to capture (e.g., does the NPI-40 really capture a person’s level of narcissism?). Construct validity can also be applied to the manipulations of variables. Do the manipulations in a study truly manipulate the variables they were designed to manipulate?
Reliability
Reliability is the consistency of a measure. If the same participants are measured multiple times under the same conditions, a reliable measure should produce similar results each time. - For a measure to valid, it must be reliable.
However, a measure can be reliable and not valid.
interrater reliability.
Interrater reliability is the level of agreement between measurements made by multiple researchers. When measurements are obtained by direct observation of behaviors, it is common to use two or more separate observers who record measurements.
internal validity
is the extent to which a research method allows us to make cause and effect inferences.
The biggest threat the internal validity
is potential confounds
External validity
is the extent to which the findings of a study generalize to other people, settings, and situations
replication
(external validity is achieved through replication) & to identify qualifiers
Replication is repeating a study with different participants, in different settings, or by using different methods. Once the results of a study have been replicated across different circumstances, it can be assumed to have higher external validity.
qualifiers
circumstances that alter the findings of a study. Typically, studies that take place in natural, real-world settings, like naturalistic observational studies, are considered more generalizable, and therefore higher in external validity.
between-participant design
each participant is only exposed to one condition of the IV and is measured once on the DV.
within-participant design
participants are exposed to all the conditions of an experiment and are measured on the DV more than once.
( don’t need to worry about random assignment)
Problems - order effects,
-Fatigue
-Practice effects
counterbalancing
must alter the order in which different participants are exposed to the different conditions).
experimental conditions.
The different levels of the IV manipulation
The placebo effect
The placebo effect occurs when people experience change from the mere expectation of change and not from the treatment/manipulation itself. There is much debate on how the placebo effect works, but it is believed to be
psychosomatic
Control group
given a placebo to control for the effect
demand characteristics
refers to any features or potential cues of a study that indicate to participants the purpose or predictions of the study
(can cause participants to try and behave in the way that they think the researchers want them to)
deception
reduces likelihood of demand characteristics from impacting results
experimenter expectancy effect.
Researchers will have expectations and hopes about the results of their study. These expectations can cause researchers to unintentionally influence the results of a
study
Double-Blind Study
the participants and the researchers interacting with the participants are blind to which condition the participants have been assigned to.
Reactivity
is when a person changes their behavior because they are aware that they are being watched and measured.