Data Flashcards
Nominal level data
When data is a headcount or frequency under a certain category
Strengths of nominal level data
Can be displayed in pie charts (+bar charts)
Quick and easy to collect data this way
Weaknesses of nominal level data
Cannot calculate a mean/median or range
Does not display a headcount of individual scores, just a headcount
How can nominal level data be displayed?
Frequency tables
Pie charts
Bar charts
Can histograms be used for nominal level data?
No
Tests to check for significance of nominal level data
Binomial sign test (if repeated measures design/ matched participants)
Chi squared test (if Independent measures)
Ordinal level data
When data is a test score/ rating score than can be put onto a scale and compare its position to the other scores
Strengths of ordinal level data
Scores for each individual is collected
Can calculate standard deviation, range, variance (dispersion)
Can calculate mean, mode, median (central tendency)
Weaknesses of ordinal data
Does not take into account the values of gaps between the scores
Using a rating scale means it could be subjective
What tests check for significance of ordinal level data?
Wilcoxon signed ranks test (if repeated measures or matched participants)
Mann Whitney U test (if independent measures)
Interval level data
Data that is placed in ranks, but the actual values of results and gaps between individual results is taken into account and known
Uses universal scales of measurement
Strengths of interval level data
We get scores from each individual
Can calculate all measures of central tendency and dispersion
Gaps between scores are known
Using universal measurements means better objectivity of results
Weaknesses of interval level data
Limited what we can measure by what universal measurements are available
Graphs for ordinal/interval level data
Scatter graph (correlation)
Line graph (change over time)
Histogram, divide continuous data into uneven groups
Primary data
Obtained by reserachers directly
Secondary data
Data that already exists from another source but was obtained by researchers to be analysed in the context of the study
Strengths of quantitative data
Easy comparisons to be made between participants/conditions
No researcher bias (can’t interpret multiple ways)
Easy to summarise (using graphs, averages etc)
Easier to establish reliability of results
Weaknesses of quantitative data
May not be representative of a participants’ everyday behaviour - low ecological validity
Doesn’t tell us why participants behave/feel that way
Construct validity issue of simplifying complex behaviour to a score, which might not be an actual measure of the score
Strengths of qualitative data
More detail about participant experience
Richer data to improve validity of results (can tell us more detail on why they behaved that way)
Weaknesses of qualitative data
Up to interpretation by researcher so subject to researcher bias
Harder to compare between participants
Cant represent on a graph
Harder to analyse using inferential statistics
Mean advantages
Involves all the data which means more representative of the average score
Mean disadvantages
Includes every score in calculation so likely to be skewed by outlier
Can be decimal
Mode advantages
Can work for qualitative data
Always a whole value
Easy to calculate
Mode disadvantages
Not always possible to calculate if a piece of data isnt repeated
doesnt include decimal points
Does not include all data collected so not representative
