Research methods: statistics Flashcards
quantitative data meaning
-data that is expressed numerically
-can be gained from individual scores such as the number of words recalled
strength of quantitative data
-quantitative data is more simple to analyse
-allows comparisons to be made been groups of data
-may be easier to make conclusions about behaviour (context)
WHEREAS
-qualitative data is wordy and more difficult to statistically summarise and therefore comparisons are hard to identify
weakness of qualitative data
-quantitative data lacks depth and meaning to behaviour
-especially when it is complex and prevents participants from being able to develop their thoughts on a given subject (context)
-therefore quantitative data may lack vital detail which reduces the internal validity of the data
WHEREAS
qualitative data is rich in detail which can provide a greater understanding of behaviour
qualitative data meaning
-data expressed in words
-may take form of a transcript of an interview or a diary entry
strength of qualitative data
-qualitative data provides rich details which allows ppts to develop their thoughts and feelings on a given subject
-this provides a greater understanding of the behaviour being studies (context)
WHEREAS
-quantitative data lacks depth and meaning as the data is only numerical
weakness of qualitative data
-harder to analyse as it is difficult to summarise to establish pattern trends
-this opens the data up to potential researcher bias as the analysis is based upon their own subjective interpretations of the data (context)
WHEREAS
-quantitative data can be analysed to provide patterns which may make it easier to make objective conclusions about behaviour
types of level of measurements
nominal
ordinal
interval
nominal data info
discrete data
data in categories
eg. hair colour
ordinal data info
discrete data
ordinal data is ranked
ordinal data does not have fixed intervals
based on subjective opinions
interval data info
data is continuous data
data is based on objective measures
based on numerical scale that have equal intervals
primary data definition
gathered first hand from the ppts
specific to the aim of the study
strength of primary data
- primary data is collected first hand from the ppts
-allows researcher to specifically target the info they require and organise an experiment in a way that suits them
strength of primary data
- primary data is collected first hand from the ppts
-allows researcher to specifically target the info they require and organise an experiment in a way that suits them (context)
-this increases the overall internal validity of the data
WHEREAS
secondary data might not meet the direct needs of the researcher suggesting it may be less useful
weakness of primary data
-primary data is conducted by the researcher themselves
-which involves time and effort to obtain the data as well as analyse the findings (context)
WHEREAS
secondary data is easily accessed and requires minimal effort to obtain reducing the time and cost taken to complete the research §6
what is meant by secondary data
-data that has previously been collected by a third party
-not specifically for the aim of the study
strength of secondary data
-can be easily accessed and requires minimal effort to obtain
-researcher might find that info she wants already exists (context)
-no need to collect primary data
WHEREAS
primary data is conducted by the researcher themselves which requires time and effort
weakness of secondary data
-secondary data may be poor quality
-may be out of date or incomplete and not met the direct needs of the researcher (context)
WHEREAS
primary data is collected first hand in ppts and specifically for the aim of the research which increases the overall internal validity of the research
what is meant by a meta-analysis
-form of research method that uses secondary data
-gains data from large number of studies which have investigated the same research
-combines the info to make conclusions about behaviour
strength of a meta analysis
gather data from a number of studies which allow us to view data with much more confidence and increase the generalizability of the findings across larger populations
weakness of a meta analysis
-prone to publication bias
-researcher may not select all the relevant studies
-choosing to leave those with negative/not significant results
-data will be biased because it only represents some of the relevant data
how to analyse qualitative data
content analysis
thematic analysis
what is content analysis
-analyse qualitative data by changing large qualitative data into quantitative
-done by identifying meaningful codes
-counted to present the data in a graph
when is it appropriate to use a content analysis
the data (context) being analysed is qualitative
what is meant by coding
initial process of content analysis where qualitative data us placed into meaningful categories
how is content analysis carried out
-read the transcript (context)
-identify codes and eg.(context)
-re-read the transcript (context)
-present the quantitative data in a graph
what is thematic analysis
-method of analysing qualitative data by identifying emergent themes
-enabling us to present the data in a qualitative format
how is thematic analysis carried out
-create a transcript (context)
-read and re-read the transcript for familiarisation
-combine the codes to create 3/4 themes that ae linked (context)
-present the data in a qualitative format
strength of analysis (content/thematic)
-easy to assess the reliability of the findings and conclusions
-researchers can access the materials and use the coding system
-to ensure the findings are consistent
weakness of analysis (content/thematic analysis)
researcher bias as the content confirms the researcher’s hypothesis is more likely to be identified and recorded
lowers the internal validity of the analysis
However
many modern (researchers) are aware of their own biases and often make reference to these in their own report
ways to assses reliability of content analysis
test re-test
inter-rater reliability
how to conduct test-retest
-research completes content analysis by creating a series of coding categories (context) and tallying every time it occurs in the qualitative data
-then the same researcher repeats the content analysis on the same qualitative data
-compare the results from each content analysis
-correlate the results from each content analysis using a stats test
-a strong positive correlation of above +0.8 shows high reliability
how to conduct inter-rater reliability
-the two raters would read through the qualitative data separately and create coding categories together (context)
-two raters read exactly the same content (context) but tally the occurrences separately
-they compare the tallies from both raters
-correlated using an appropriate stats test
-a strong positive correlation shows high reliability +0.8
definition of operationalisation
be specific and clear when defining coding categories to make the codes more measurable
how to assess the validity of content analysis
face validity
concurrent validity
face validity process
-independent psychologist in the same field seeing ig a coding category (context) is measuring what it claims to measure (context) at first value. If they say yes the content analysis is valid
concurrent validity process
-compare the result of a new content analysis (context) with the results from another similar pre-existing content analysis which has already been established
-if the results from both are similar then we can assume the test is valid
-correlation of two sets of results gained from an appropriate stats test should exceed +0.8
what is meant by measures of central tendency
any measure of the average value in a data set eg. mean
mode AO1
Most common number in a set of scores
Can be more than one mode of a data set
-Used for nominal data
mode advantages
-easy to calculate
-less prone to distortion by extreme values as it does not take all data in to account
UNLIKE
the mean which is highly influenced by extreme scores
mode disadvantages
-does not take account of all scores
-not useful if more than one mode
median ao1
-middle score in ordered scores
-ordinal data
median advantages
easy to calculate
not affected by extreme values
median disadvantages
not as sensitive as mean as does not use all scores
mean ao1
-all scores added up and divided by the total number of scores
-used for interval data
mean advantages
-most accurate and sensitive measure as uses all scores