research methods Flashcards
define content analysis
indirect observation of behaviour within the media and can produce qualitative data and quantitative data
stages of content analysis
data is collected, researcher reads and examines data, identify coding units, apply coding units, tally, theme
2 strengths of content analysis
less ethical issues, produces quantitative and qualitative data
limitation of content analysis
subjective
define case studies
in depth investigation, description and analysis of a single individual or small group
define longitudinal studies
case studies that take place over a long period of time and may involve gathering data from family and friends
2 strengths of case studies
offer rich in depth detail increasing validity, generate hypothesis for future study
2 limitations of case studies
can’t generalise, low in validity (subjective)
define reliability
study or test result should be consistent and be able to be reproduced
define external reliability
how consistent/reliable a measuring instrument is across different times and situations
define test retest
same test/study is conducted with the same people on more than one occasion and checked for a significant positive correlation
test retest steps
plot the 1st test, plot the 2nd test, see if there’s a significant correlation, Spearman’s statistic test
limitation of test retest
practice effects
define inter-observer reliability
having 2 observers making recordings of same behaviour, at same time but independently. compared to see if they’re similar
define validity
does the test measure what it claims to measure
define face validity
on the face of it, does the test look valid
define concurrent validity
group of people are given a new test and then an established measure of the same ability
define ecological validity
extent to which findings can be generalised
define temporal validity
extent to which findings can be generalised to other historical times and eras
improving reliability - questionnaires
ambiguous or complex questions removed, replacing open questions with closed questions
improving reliability - interviews
using same interviewer to avoid leading questions
improving reliability - experiments
standardised instructions and procedures
improving reliability - observation
operationalisation and explanation of behavioural categories
improving validity - questionnaires
lie scale to control effects of social desirability bias, anonymous
improving validity - qualitative methods
interpretive validity, triangulation
improving validity - experiments
control groups, single blind and double blind procedures
improving validity - observations
covert observations, behavioural categories operationalised
analysing data
decide level of measurement, decide descriptive statistics, choose statistical test
levels of measurement
nominal, ordinal, interval
define nominal data
categories
weakness of nominal data
least sensitive measure, only use the mode
define ordinal data
rank or position
weakness of ordinal data
doesn’t show actual difference, very subjective
strengths of ordinal data
more information than nominal, uses median, mode and range
define interval data
precise scales
strengths of interval data
most sensitive, uses mean, mode, median, range, standard deviation
measures of central tendency
mean, median, mode
define mean
average
strengths of mean
most information, most sensitive
mean limitations
greatly affected by extreme scores
define median
central number
median strengths
not affected by extreme scores
median limitations
not as sensitive
define mode
most common
mode strengths
not affected by extreme scores
mode limitations
rarely useful in small sets
define bimodal
two modes
measures of dispersion
range, standard deviation
define range
spread of a set of scores
range strengths
easy to calculate and understand
range limitations
distorted by extreme values, doesn’t say if scores are clustered or not
define standard deviation
how far each score is away from mean
standard deviation strengths
all scores are used, more sensitive
standard deviation limitations
time consuming
spearmans rho
correlation, ordinal data
Pearson
correlation, interval data
chi squared
difference, unrelated, nominal data
mann whitney
difference, unrelated, ordinal data
unrelated t test
difference, unrelated, interval data
binominal sign test
difference, related, nominal data
wilcoxon
difference, related, ordinal data
related t test
difference, related, interval data
unrelated design
independent measures design
related design
matched pairs or repeated measures design
define significance level
probability that results have occurred by chance
1% significance level
medical trials and previous research. 99% sure results are due to test, not due to chance
5% significance level
everyday psychological testing. 95% sure results are due to test, not due to chance. reduces type 1 and type 2 error
10% significance level
pilot study. 90% sure results are due to test, not due to chance
type 1 error
when we say results are significant but they aren’t significant. reject H0, accept H1
type 2 error
when we say results aren’t significant but they are significant. reject H1, accept H0
define histograms
shows data is continuous
scientific approach in psychology aims
describe, understand, predict and control behaviour
define paradigm
shared set of assumptions and agreed methods within a scientific discipline
SLT - physical observes, experiments
psychodynamic - inference
define paradigm shift
result of a scientific revolution; significant change in the dominant unifying theory within a scientific discipline
biological/biopsychology - new medicines
define theory construction and hypothesis testing
theory should be testable so it can be supported or falsified. to do this, we arrive at a hypothesis
SLT - children learn through imitating and observing role models
psychodynamic - unconscious mind controls behaviour
define falsifiability
principle that a theory can’t be considered scientific unless it admits the possibility of being proven untrue
SLT - conducting experiments
psychodynamic - inference
define empirical method
scientific approaches that are based on the gathering of evidence through direct observation and experience
SLT - experiments
psychodynamic - no experiments
define objectivity
when all sources of personal bias are minimised so as not to distort or influence research process
SLT - control group and standardisation
psychodynamic - case studies
define replicability
extent to which scientific procedures and findings can be repeated - can they be repeated now and in 20 years
SLT - yes
psychodynamic - no
define confidentiality
participants data and results remain confidential
define lack of confidentiality
participants data has been released without consent
define deception
participants shouldn’t be lied to about true aim of research
what must the researcher ensure if deception is essential
there’s no other way to carry out the procedure, debrief participants, consult with colleagues
define right to withdraw
participants have right to withdraw at any point, payment doesn’t alter this
define lack of right to withdraw
participants feel like they must continue and they can’t leave the study
define informed consent
researcher must attempt to get real consent from participants. can only be possible when they full understand the study
consent for children
obtained by parents
consent for adults with impairments
obtained by someone who would know if the person would like to take part
define lack of informed consent
participants haven’t been told enough about the study to make an informed decision
define protection from harm
participants should be protected from psychological and physical harm
define brief
said to participants before the study
consent form
participants should be issued with consent letter or form