Data handling Flashcards
Social change
understanding social influence helps produce campaigns to improve behaviour
e.g correcting misperceptions so the norm becomes better e.g Montana reduced alcoholism by saying only 20% of people drink drive reducing the majority thought that most people do that
or using conformity to prevent undesirable behaviour
hence reducing burden/cost on e.g health services
Improving memory
Cognitive interview to improve eyewitness accounts, less money spent on wrongful arrests and wasted police resources
Attachment
Research developed since traditional monotropic bond theory, now both parents can care for kids and both parents can work flexible arrangements
Maximising income and contributing to economy
Knock on effects like paternity leave funded by government and affecting companies and gender pay gap decrease
Mental Health
direct costs of mental health e.g social care cost 22.5 billion a year, and 15 billion a year for absences from work for issues like stress and depression
Research in improving drug therapies and others improve mental health so less absences and money lost
Help people manage lives productively
Cutting edge research findings can attract overseas funding
But new treatments can burden NHS financially, and may be more effective but more expensive too
Types of data
Nominal - discrete data in separate categories e.g eye colour, one person can only be in one category
Ordinal - continuous data with some sort of ordering or ranking e,g listing music genres in rank order
Interval - continuous data in equal intervals, or using own scales e.g scale of 1-10 with arbitrary units
Quantitative data
Numerical data e.g how much, how long e,g tallies or closed questions
+ Easy to analyse and test, draw patterns and comparison, more objective and unbiased data
- Lacks validity and might not fully address key aim variables, lacks detailed meaning, tells us the what but not the why
Qualitative data
Unquantifiable, usually lengthy and detailed e.g reports, pictures, open questions and observation notes
+ Highly detailed to see complexity of behaviour, high validity and assess concepts and ideas linked to aim, and easy to establish cause effect relationship with IV and DV
- Unreliable as repeats will have different results, subjectivity in analysis and behaviours so hard to make valid conclusions
Data collection types
Primary - first hand collection, exactly what is required but requires time and effort
Secondary - stats or data that is already collected for other purposes but links to the aim - easy to source but reliability and validity questionable
Meta-analysis - combining results from multiple studies and reviewed together - reliable and able to generalise with confidence, but risk of publication bias where studies with undesirable results are ignored
Data presentation
Bar charts for nominal discrete data
Histograms used for interval (or ordinal) data
Line graphs for ordinal or interval data
Scattergraphs - bivariate correlations
Make sure to have a clear title with x and y variables stated
Distributions
Most data sets have normal distributions will bell shaped curve
positive skew has most data on the left and some stretched to the right, so mode is at peak, median a bit after and mean after that
vice versa for negative skew
Measures of central tendency
mean - most sensitive and accounts for all scores, but can be distorted by extreme values and is not one of the actual scores so could be ‘impossible’ e.g 5.5 people
median - unaffected by extreme values but less sensitive and not good if most data is clustered at high or low levels
mode - unaffected by extreme scores but tells us nothing about the other data
Measures of dispersion
Range - very quick and easy to calculate but most affected by extremes
Standard deviation - accounts for all scores but difficult to calculate and only works for interval/ nominal data
Inferential statistics
Inferring something for a whole population based on a sample
Changed inferences can be made only when research shows significant data (i.e too lucky to be by chance)
generally 5% in psychology but 1% for things like new medicine
When to use sign test
Looking for difference between data, paired or related data i.e matched pairs or repeated measures as treated as one person tested twice, and nominal (quantitative) data
key word ‘before and after the…’
How to sign test
Make a hypothesis (1/2 tailed), then work out the sign change for each participant by doing experimental score - control score, then total up number of + and - where test statistic S is lower value
N is number of participants but ignore 0 scorers
Then match up critical value where S needs to be equal to or less than for it to be significant
