Research methods Flashcards
3 main types of extraneous variables
1) participants variables- personal characteristic of the participant
2)investigator effects-when the researcher influences the behaviour of the participant, this can be the result of a researchers behaviour/personality or appearance
3) situational variables- features of the external environment that can affect the results of a a study
demand characteristics
-are when situational and experimental variables/investigator effects act as hints that enable participants to guess the aim of the study
demand characteristics: influence on behaviour
demand characteristics can influence behaviour in 2 ways:
1)some participants try to be helpful and behave in the ways they think they’re suppose to behave
2)some participants might rebel and behave the opposite to how they think they’re suppose to behave
effect of extraneous variables
-uncontrolled extraneous variables reduce both reliability and validity of a study
standardisation
-when researchers make an extraneous variable the same for all participants in a study
-used to control situational variables and experimenter variables/investigator effects
controlling extraneous variables: matching
matching is when a particular characteristic of the participants is divided equally across groups
controlling participant variables: random allocation
-when participants are assigned at random to experimental groups
controlling demand characteristics: blinding
-there are 2 types of blinding:
single blind study:
-when the participant doesn’t know which experimental group they are in
double blind study:
-when neither the participant or the researcher know which experimental group the participants are in
-this can control both investigatory effects and demand characteristics
the self-report technique
-when participants are asked to provide information about their own thoughts, feelings and behaviours
2 types of self report:
interviews:
-questions are asked directly, face to face to the participant
questionnaires:
-the questions are written down and the participants have to write down their response
2 types of interviews:
structured:
-all the questions are decided in advance
unstructured:
-not all questions are decided in advance and the interviewer may decide on additional questions based on the participants response
types of questions in questionnaires:
open questions:
-questions where participants are asked to answer in their own words
closed questions:
-questions wheres participants Re given a range of fixed answers to chose from
limitations of self-report technique:
-people aren’t always accurate when they provide information about how the feel,think or behave
-people may give answer that they think are socially desirables to avoid being judged negatively (social desirability bias)
questionnaires vs interviews
-questionnaires are quicker to collect data because questionnaires can be more easily and sickly distributed to a large number of participants, whereas with interviews, researchers need to spend time with watch participants w by one
-in questionnaires, participants are less likely to be influenced by investigator effects
-data collected by interviews have higher quality as researches can make sure that all of the questions are properly understood and answered
and in an unstructured interview, researchers came also ask more questions based on what the participants have responded,meaning they can get more detailed information
-interviews may lead to a less bias sample than questionnaire
strengths of closed questions:
-responded to closed questions are easier to count up and analyse compared to open questions
-easier and after for participants to answer, making it easier to recruit participants for the study, this is likely to make the study more representative
strengths of open questions:
-answers are more likely to be representative of the participants actual attitudes and feelings l, which makes the results more valid
-researchers are likely to get more detailed answers using open questions
evaluation of structured interviews:
strengths:
-more reliable than instructed interviews
limitations:
-however, unstructured interviews allow the researcher to obtain more information than structured interviews
event sampling
-when researchers focus on normal specific behaviours and count to every time those behaviour occur
observational technique
-when researchers observe participants and measure or record their behavior
3 key decisions that researchers have to make when using the observational technique
1)type of observation to conduct:
-controlled observation is when researchers conduct observations on pps in controlled, artificial experiments
-a naturalistic observation is an observation conducted in an everyday life setting
2)participant awareness- researchers need to decide how aware their pps are of their observations:
-in an overt observations, pps are aware they are being watched whilst in a covert observation pps are not aware they are being watched
3)researcher role-researcher needs to decide what their own role will be in the study:
-a participant observation is when the researcher participates in the activity the pps are doing whereas a non-participants observation is when the researcher doesnt take part in the activity the pps are doing
time sampling
-when researcher categorises behavior at regular intervals, like every 30 seconds
evaluation of observational techniques
-controlled observation have better control over extraneous variables than naturalistic observations, but natural observations have higher ecological validity than controlled observations
-overt observations are more ethical than covert observations because researchers can obtain informed consent, but covert observations are less likely to be effected by investigators effects or socials desirability bias
-in participants observations, the researchers may gain more insight and understanding into the behaviour of their participants, but non-participants observations are less effected by investigator effects
weaknesses of observational techniques:
-doesn’t tell us about people’s thoughts and feelings as u are only recording behaviours
-observational technique may lead to observer bias, which is the tendency for researchers to see what they expect to see when conducting observations , making the observation less accurate and objective
2 ways to reduce observer bias in an observation:
1) researcher should break down the behaviour they are recording into behavioural categories
2)researchers can calculate inter-rated reliability, this is a measure of how similar the data collected by different observers are and it assesses the external reliability of observations in an observational study
Pilot studies
-a pilot study is a small test study conducted with a small number of pps to identify apotential flaws in an experiment before the proper experiment beginns
-this is different from a caste study which is a study of unusual or rare experiences that couldn’t be studied in an experiment
-also, researchers don’t analyse results from a pilot study, because the design of the study might change and the small sample size means the results wouldn’t be very reliable anyway
benefit of conducting pilot studies
-they are cost effective, so you don’t waste time and money on big experiments and end up with inaccurate results
Peer review
-the process where other researchers in the field review, criticise and suggest improvements for a report before it’s published
Evaluation of Peer reviews
strengths:
-peer review protects society from bad research
-helps researchers to constantly improve their research
limitation:
-the reviewers might not always be very objective, this is because it is not always in their interest to be objective and fair, and even when reviewers try to be fair, they might display personal bias, this means that good studies might never get published, meaning that society misses out
types of data: qualitative & quantitative
qualitative-can’t be described using numbers
quantitative-can he described using numbers
4 levels of measurement/ types of quantities data: nominal data
-data that is grouped into categories, where the categories don’t have a natural order
4 levels of measurement/ types of quantities data: ordinal data
-data that’s grouped into categories that do not have a natural order
4 levels of measurement/ types of quantities data: interval/ratio data
ratio data- data that tells us exactly how much bigger one number is than another but who’s values are numbers that can’t go below zero e.g. mass
interval data- data that tells us exactly how much bigger one number is than another but who’s values are number that can go below zero e.g temperature
quantitative data: discrete and continuous data
discrete- quantitative data that’s restricted to just certain numbers, nominal and ordinal data are always discrete
continuous data-quantitative data that’s not restricted to jus certain numbers
-ratio or interval data can be rather continuous or discrete
frequency graph
-the X axis displays continuous data and the Y axis displays frequency count
-used to represent frequency counts for continuous data
what is data distribution:
-the shape of the curves is called the distribution of data
-a curve that is symmetrical around the middle and shaped like a bell curve is called a normally distributed curve
-when the curve is centred to the left and has a long tail on the right, we say it has a positive skew
-whereas, when the curve is centred to the right and has a long tail on the left, we say it has a negative skew
central tendency:
-a number that tells us where the middle of the distribution is located
-we can measure the central tendency in 3 ways:
-mode
-mean
-median
measure of central tendency and skew
-if data is normally distributed, the mode, mean and median are all located in the same place
-but, on a positively skewed distribution, the mode is located at the peak, the median is to the right of the mode, and the mean is to the right of the median
-on a negatively skewed distribution, the mode is located at the peak, the median is to the left of the mode, and the mean is to the left of the median
measuring dispersion:
-tells us how spread out the distribution is
there are 2 ways we can measure this:
1)range- the difference between the highest value and the lowest value
2)standard deviation-measures the average distance of the data from the mean of the distribution
