research methods- y13 Flashcards
what are case studies
-detailed in depth and longitudinal
-tend to take place over a long period of time
-may involve data from family and friends
-unusual or typical cases
-usually qualitative data
-may inc quantitative in intelligence or personality
evaluation- pros and cons of case studies
pros
-rich and detailed inc validity
-enables study of unusual behaviour
cons
-prone to researcher bias making subjective conclusions which reduce validity
-small samples, difficult to generalise and low external validity
content analysis
-systematicaly summarising data to produce quantiative data -people studied indirectly through their communications
-may inc; spoken interaction, written forms or media
-coding may prod quantitative data
-info categorised into meaningful units
-includes thematic analysis
-tally categories
-check reliability by correlating scores
thematic analysis
-produces qualitative data
- aims to produce themes
-themes are more descriptive
- uses stereotyping
-uses quotations
-data is transcribed, look for recurring patterns, provide examples to illustrate themes
evaluation- pros and cons of content analysis
pros
-ethical issues may not apply
-flexible method
cons
-communication studied out of context reducing validity
-may lack objectivity and threaten validity
what is reliability
-if a particular measure is repeated and same result obtained then that measurement is described as being reliable
-consistency
Ways of assessing reliability
1.test- retest= test same person twice
2.inter observer= 2 or more observers compare data by conducting pilot study, record data independently
3.reliability measured using correlation= correlation coefficient exceed +0.8 for reliability
ways of describing data in bar charts
-discrete
-nominal
-categorical
what does operationalised mean?
-measurable
how is range calculated
highest score-lowest score +1
ways of improving reliability
1.questionnaires- rewrite questions
2.interviews- improve training
3.observations-categories operationalised and no overlap
4.experiments- standardised procedures
what is validity?
-when another observed effect is genuine and represents what is actually ‘out there’ in the real world
-data can be reliable but not valid
types of validity- 4
1.internal- control within to reduce demand characteristics
2.external- generalise to other settings
2a.ecological- generalise to other settings in everyday life
2b.temporal- consistent over time
ways of assessing validity- 2
1.face validity- test looks like it measures what it claims to measure, eyeballing
2.concurrent- whether findings similar to those in well established test
ways of improving validity
-control groups and standardisation in experiments
-ensure changes in dv were due to manipulation of iv
-lie scale
what is a statistical test?
-used to determine whether a difference or association/correlation found in an investigation is statistically significant
-i.e. whether it could have occurred by chance or a real effect
three criteria for a statistical test
1.looking for difference or association
2.is it related (repeated measures/matched pairs) or unrelated (independent groups)
3.what is the level of measurement
table of choosing a statistical test
test of difference association
unrelated related nominal data chi-2 sign test chi-2
ordinal data man whit wilcox spear
interval data unrelatedt relatedt pears
levels of measurement/ types of data
1.nominal
-categories
-each item can only be in 1 category
-e.g. fav football team
2.ordinal
-place in order with subjective intervals
-numerical ordered scale
-e.g. rating happiness
3.interval
-units of equal size
-numerical scale
-more detail preserved
probability and significance
-null hypothesis states that there is no difference or no correlation
-if it is not significant the null hypothesis is accepted
-probability is a measure of how likely it is that a particular event will occur where 0 is a statistical impossibility and 1 is certainty
use of statistical tables
-calculated value compared with critical
-hypothesis can be one or two tailed
-number of pps is degrees of freedom
-level of significance/ p value
-usual significance level is 0.05 or 5%
-more stringent 1% can be used for drug trials
type 1 error
-null hypothesis rejected and alt accepted when null is true
-false positive, showing significance when it does not exist
-more likely if significance level too lenient e.g. 10%
type 2 error
-null accepted but alt is actually true
-false neg/ pessimistic error
-more likely if significance level too stringent or low e.g. 0.01%
sections of scientific report- abstract
-short summary, 150-200 words
-includes all major elements; aims hypothesis , method, results and procedure
sections of scientific report- introduction
-literature review
-look at relevant theories
-should follow logical progression beginning broadly and becoming more specific until aims and hypothesis presented
