Research Methods Flashcards
population
the entire group that is of interest to the researcher
sample
sub set of the population
convenience sampling
sampling technique where subjects are recruited from a nearby source due to convenient accessibility and proximity to researcher
random sampling
sampling technique where population is given equal chance of being selected to be part of the sample
stratified random sampling
sampling technique where population is divided into group and random sampling is done within each group
snowball sampling
sampling technique where individuals of interest are interviewed and they would provide names of others of interest who may consent to being interviewed for the study
experimental method
involves manipulation of IV to see its effects on the DV to establish cause and effect
independent variable
variable manipulated by experimenter to see its effects on another variable
dependent variable
variable measured in response to changes in manipulated variable
experimental group
group of participants subjected to manipulation of IV
control group
group of participants who are not exposed to IV, rather used to provide comparison to the experimental group
controlled variable
variable that remains the same for both experimental and control group throughout the experiment so that the changes observed are the results of the IV
uncontrolled variable
variable that is allowed to stay random as it is predicted to have no effect on the outcome of the experiment
hypothesis
scientific statement based on the available information that can be tested by experimentation
- when appropriate, the statement expresses an expected relationship between IV and DV for the observed phenomena
placebo
dummy pill or harmless substance given as if it were treatment
- used to control participants expectations
variables that can affect changes in DV and outcome in IV
- personal characteristics
- desire to look good or perform appropriately
- demand characteristics: cues given by experimenter of what is expected of the participants
reliability
extent to which the researcher would obtain similar results when the same test is given to the same person again under the same circumstances
types of reliability
- internal consistency: extent to which the results across items within a test are consistent
- split half method: where scores on one half of a measurement are correlated to the other half of the measurement
- test retests reliability: where participants should obtain similar results across time by repeating the same test on different occasions
validity
extent to which the results of a test reflects what the measurement instrument says it is measuring
types of validity
- face validity: when a measurement looks like what it is supposed to measure
- construct validity: to see whether test items are keeping with the constructs on which the test was based on
- concurrent validity: where scores on a measurement are correlated with another measurement that reflects the same construct
- predictive validity: extent to which test results predict other behaviors related to the constructs tested
study design
set of methods and procedures used in collecting and analysing measures of the variables specified in the research problem
cross sectional study design
research design which takes into account age related developmental changes by comparing groups of different ages at the same point in time
- benefits: time efficient, cost efficient, researchers are less likely to lose their participants
- limitations: does not provide for changes over time, cannot eliminate cohort effect
longitudinal study design
research design which follows developments of one group over time to look for changes
- benefits: eliminates cohort effect, allows for study of actual changes with development
- limitations: expensive to follow group for long period of time, researchers run risk of losing participants, participants may experience practice effect
longitudinal-sequential study design
research design which combines features of both cross sectional and longitudinal design in an attempt to overcome the limitations of each design
- groups of participants are followed over time and at every measurement point, a new group is added that is the same as the first group at the first measurement point
- benefits: allows for actual changes over time, allows to look for age differences in behavior, check for cohort effect
- limitations: complex, expensive, requires well organised research team to track data
non experimental method
does not involve any manipulation of variables
observation
study of people by observing them
types of observations
- structured observation: where researchers gather data without getting directly involved with the participants
- naturalistic observation: observation of subjects in their natural environment
case study
study which provides detailed knowledge about a single case or small number of related cases
survey
self reported data from participants who ideally have been randomly selected
types of surveys
- fixed response surveys: consists of questions which usually consists of yes or no answers or rating a response from 0 to 10
- open ended surveys: consists of questions which will ask participants to comment freely on a statement or question
correlational study
study which investigates whether there is a relationship between variables and how strong that relationship is
archival research
study which involves examination of old material usually found in places like libraries or government buildings
qualitative method
- data cannot be expressed in numbers and cannot be analysed statistically
- benefits: provides in depth information, flexibility of explanations in participants responses, explores new area of research
- limitations: requires a lot of time to sift through detailed information, cannot be generalised to population, responses might be affected by social desirability bias
social desirability bias
peoples tendency to behave in a certain way that they consider socially acceptable even if they don’t normally behave that way
interview questions
- open ended interview: researchers ask participants to comment freely on statements or questions in any way they like
- fixed response interview: responses are restricted to a range of options offered by researcher
quantitative method
data can be expressed in numbers and analysed statistically
objective quantitative method
measures are generally physiological responses
- advantage: difficult for participants to manipulate responses
subjective quantitative method
- responses are provided on a checklist or likert scale to measure a range of abilities, attitudes, views and opinions
- advantages: responses provides in depth information
- disadvantage: responses might be affected by social desirability bias
measures of central tendency
identifies location of centre of various distributions
- mean: average score
- median: middle score
- mode: most frequently occurring score
measures of dispersion
finds out how spread out data values are on the number line
- range: difference between the largest value and smallest value
- variance: spread of scores around the mean
- standard deviation: average amount by which scores differ from the mean
bell curve
display of distribution of scores
features of bell curve
- area under the normal curve is always equals to 1
- mean, mode and median are identical
- defined according to mean and standard deviation
left skewed
when most of the data is concentrated on the right end of the curve
right skewed
when most of the data is concentrated on the left end of the curve
statistical significance
value used to indicate whether differences between the mean scores in sets of data for groups being studied are real or due to chance
p > 0.05
- results are due to more than 5% of chance or random error.
- results do not have statistical significance
- there is no significant difference in DV between EG and CG
- hypothesis is rejected
p < 0.05
- results are due to less than 5% of chance or random error
- results have statistical significance
- there is a significant difference in DV between the EG and CG
- hypothesis is accepted
single blind study
experimental procedure in which the researcher is aware of whether participants are in EG or CG, however participants are blind to their membership
- prevents participants from displaying social desirability bias
- prevents placebo effect
double blind study
experimental procedure in which both researcher and participants are blind to whether participants are in EG or CG
- prevents experimenter’s effect
experimenter’s effect
experimenter’s values and beliefs which can influence the research process at any point
professional conducts
- dressing appropriately
- being punctual
- using appropriate language
privacy
right of protection from unwanted intrusion by government or other people into one’s affairs
confidentiality
degree of secrecy attached to information provided by client to researcher
- information should not be disclosed to anyone but the researcher unless given consent
- ensure data is stored securely in locked cabinet or password protected computer
- destroy data securely
- ensure that there is no identifying information that may be linked to client
anonymity
protection of people’s identity through not disclosing their name or not knowing it
- researcher should not ask for any identifying information from participants
- use code to identify participants
- store participants identifying information separate from their research data
voluntary rights
participants must willingly decide to take part in the experiment without any persuasion or coercion
- participants should not receive any threats or negative consequences if they decide not to participate
withdrawal rights
participants have the right to leave the research at any time without any negative consequences
informed consent
approval of participant to participate in a research based on their knowledge of why procedure is being carried out and what is involved
- researcher must ensure participants are aware of the nature and objectives of the study, duration, what they will be expected to do and the potential risks and benefits that can be obtained in a language they can understand
- minors must obtain consent from their parent or legal guardian as they are considered to not be able to understand what they are being told
deception
necessity of researcher to hide the real reason behind the research in order to reduce the likelihood that participants will behave differently
- researcher must debrief participants after the experiment and explain why deception has been used
beneficence
where the benefits must outweigh any potential risks obtained from an experiment
