Data Handling and Analysis Flashcards
what is quantitative data?
quantitative and qualitative
data that is in numerical form. e.g. reaction time, tally of times behaviour is shown, rating out of 10.
what is qualitative data?
quantitative and qualitative
data that is in the form of words. e.g. descriptions of behaviour, feelings or emotions
how can qualitative be turned to quantitative?
quantitative and qualitative
operationalise the behaviour (behavioural categories) and then tally.
how is quantitative data collected?
quantitative and qualitative
experimental and observational techniques, closed questionnaires.
how is qualitative data collected?
quantitative and qualitative
case studies, open question interviews and questionnaires
what are the strengths and limitations of quantitative data?
quantitative and qualitative
S: objective so no bias, descriptive statistics allow data t be summarsied in graphs, more reliable
L: lacks depth and detail, only focus on behaviours that can be mathematically tested.
what are the strengths and limitations of qualitative data?
quantitative and qualitative
S: rich in depth and detail to provide high understanding + validity
L: subjective so risk of bias, challenging to summarise, less reliable
what is primary data?
primary and secondary data
data collected first hand by the researcher, specifically for the purpose of the investigation
what is secondary data?
primary and secondary data
data that is collected second hand from already published sources that didn’t initally set out to answer the current research question.
what are the strengths and limitations of primary data?
primary and secondary data
S: high validity as data collected to answer Q specifially and researcher has control of collection process.
L: requires time and effort to develop resources, may be costly compared to secondary data which can be easily accessed.
what are the strengths and limitations of secondary data?
primary and secondary data
S: already exists so very quick
L: low validity as not collected initially to answr Q and researcher has no control over collection process (variables)
what is a meta analysis?
primary and secondary data
collecting and combining the results from previously conducted studies investigating a similar question and analysing them.
what are the strengths and limitations of a meta analysis?
primary and secondary data
S: large sample size, more trustworthy as extraneous influence reduced
L: secondary data, choice of which studies are or arent included leads to bias, statistically unimportant studies not included
what are descriptive statistics?
descriptive statistics
describe and summarise data collected.
what are the two brances of descriptive statistics?
descriptive statistics
- measurs of central tendency
- measures of dispersion
what is a measure of central tendency?
descriptive statistics
a single value that summarises a set of data, identifying a typical value (the average)
- mean
- mode
- median
what is the mean?
descriptive statistics
the arthimetic average of a set of data, calculated by adding all values and dividing by number of values.
what are the strengths and limitations of using the mean?
descriptive statistics
S: all raw data used so most sensitive measure
L: easily disrorted by extreme scores (outliers)
what is the mode?
descriptive statistics
the most frequently occuring value in a set of data.
if 2 - bimodal
if more than 2 - multimodal
what are the strengths and limitations of using the mode?
descriptive statistics
S: unaffected by outliers, only way to give average in categories (e.g. pet type)
L: doesn’t take all values into account, unreliable in small data sets (no mode if all values are the same)
what is the median?
descriptive statistics
the middle value of a set of scores by putting values into numerical order and selecting middle. if even, halfway between the middle data points.
what are the strengths and limitations of using the median?
descriptive statistics
S: easy to calculate, unaffected by extreme outliers.
L: doesn’t include all values in the data set, if even set, the median won’t be a value in the data set at all.
what are measures of dispersion?
descriptive statistics
values that summarise the spread of the data.
- range
- standard deviation
what is the range?
descriptive statistics
the difference between the highest value and the lowest value in a set of data.
what are the strengths and limitations of using the range?
descriptive statistics
S: easy to calculate
L: easily distorted by extreme values, doesn’t represent every value in the data set
what is the standard deviation?
descriptive statistics
a measure of how much each score deviates from the mean on average.
what is the equation to calculate standard deviation?
descriptive statistics
sum of (value - mean) squared divided by number of values - 1. all square rooted.
what does the calculated standard deviation value mean?
descriptive statistics
the larger the SD, the more spread out the data is around the mean -> less representative and less reliable.
what are the strengths and limitations of using the SD?
descriptive statistics
S: uses all values so more sensitive measure of spread
L: more difficult to calculate, distorted by extreme results
how are percentages used in psychology?
descriptive statistics
https://www.youtube.com/watch?v=TsSWLB0aEWs&list=PLUQ8QDGvbAwhFY-fZkcJ3k4R2NCnZlqB4&index=19
what are the three types of tables that can be used to display data?
displaying quanitative data
- raw data table
- frequency table
- descriptive statistics table
what is a raw data table?
displaying quanitative data
a record of individual data points.
what is a frequency table?
displaying quanitative data
a tally chart, log of the number of times a behavioural category is seen
what is a descriptive statistics table?
displaying quanitative data
a table with the measures of central tendencies and measures of dispersion of behavioural categories.
what is a bar chart?
displaying quanitative data
a chart that summarises the frequency of a category
what type of data does a bar chart display?
displaying quanitative data
nominal data (categories). e.g. type of pet
what variable goes on which axis on a bar chart?
displaying quanitative data
X - category
Y - frequency
what does the height of the bar represent in a bar chart?
displaying quanitative data
the frequency
what is important when drawing a bar chart?
displaying quanitative data
LINES DO NOT TOUCH AS DATA IS NOT CONTINUOUS - THATD BE A HISTOGRAM
what is a pie chart?
displaying quanitative data
a circular graph that represents all the data in the set. each wedge represents the proportion of one category.
what type of data is displayed in a pie chart?
displaying quanitative data
nominal data -> each category has a ‘wedge’ where its size represents the proportion
what does a scattergram display?
displaying quanitative data
the relationship between two co-variables. (correlations)
what is on the x and y axis of a scattergram?
displaying quanitative data
the co variables.
e.g X - attendance
Y- test score
what does each point on a scattergram represent?
displaying quanitative data
one participant’s measure. e.g. their attendance and their test scores.
what can a scattergram indicate?
displaying quanitative data
a positive or negative correlation between the co variables.
what is a histogram?
displaying quanitative data
a graph that displays the frequency of continuous, numerical data.
what variable is placed on which axis in a histogram?
displaying quanitative data
X - continuous variable e.g. age group, months etc.
Y - frequency
what is important when drawing a histogram?
displaying quanitative data
THE BARS ARE TOUCHING AS IT IS SHOWING CONTINUOUS NOT CATEGORIAL DATA.
what is a line graph used for?
displaying quanitative data
to display two sets of continuous data on the same graph.
how is a line graph drawn?
displaying quanitative data
place an X on the midpoint of the top of each bar and join with a line.
what is a line graph also known as?
displaying quanitative data
a frequency polygon
what variables go on which axis on a line graph?
displaying quanitative data
X - continuous variable
what graph is used in distributions?
distributions
histograms
what is a normal distribution
distributions
when the frequency distribution forms a symmetrical bell shaped curve
what does a normal distribution curve indicate?
distributions
more ppts are in the middle, with fewer on either side
where is the mean, mode and median found on a normal distribution curve?
distributions
the top at the centre of the curve
where is the mode found on a normal distribution curve?
distributions
the top of the curve as most frequent.
where is the median found on a normal distribution curve?
distributions
the middle score
where is the mean found on a normal distribution curve?
distributions
the middle as an equal number of outliers on either side
why does the normal distribution never touch the x-axis?
distributions
because there will always been an extreme score from at least one person.
what is standard deviation like on a normal distribution curve?
distributions
68% fall within one SD of the mean, 95% within two SD and 99.7% within three SD.
what is a skewed distribution?
distributions
when the distribution of the scores are assymetric - most scores on one side.
what does a posiitve skew look like?
distributions
the ‘tail’ points to the right, with the majority on the left.
what does a negative skew look like?
distributions
the ‘tail’ points to the left, with the majority on the right.
how is data distributed on a positive skew?
distributions
more scores at the lower end, outliers at higher end of x axis.
how is data distributed on a negative skew?
distributions
more scores at the higher end, outliers at the lower end of x axis.
where is the mode found on a skewed distribution?
distributions
the highest point still
where is the median found on a skewed distribution?
distributions
at the point where 50% of graph is either side.
where is the mean on a skewed distribution?
distributions
shifted towards the outlier scores
what are the 3 levels of measurement?
levels of measurements
- nominal
- ordinal
- interval/ratio
what is nominal data?
levels of measurements
categorical data that has no natural order - the frequency count of the variable is recorded. e.g. birth country, pets, career choice
what are the variables in nominal data?
levels of measurment
discrete (non overlapping).
what is ordinal data?
levels of measurment
categorical data that has a natural order - it can be placed into some kind of order or scale
is the difference consistent on ordinal scales?
levels of measurement
no
what are examples of ordinal data?
levels of measurement
positions in a competition, height in a class, choice on a likert scale.
what is interval data?
levels of measurement
data that is measured in fixed units with equal distance between points on the scale. e.g. temperature, a mm ruler - precise and continuous
what is ratio data?
levels of measurement
interval data with an absolute zero e.g kelvin scale
how do you convert interval to ordinal data?
levels of measurement
- gather the interval data
- list each ppt’s data from highest to lowest
- assign a rank
how do you convert ordinal to nominal?
levels of measurement
- create categories (fast and slow)
- place highest half into one category and lowest into the other.
how do you perform content analysis?
content analysis and coding
- decide on a research question
- select a sample from all possible content
- coding : decide on categories (coding units) to be recorded
- work through data - tally number of times category appears
- data analysis
what is a coding unit in content analysis?
content analysis and coding
the operationalised categories decided beforehand that will be looked for in the content analysis
how do you test for reliabaility of a content analysis?
content analysis and coding
- test-retest reliability: run the analysis again on the same sample and compare
- inter-rater reliability: a second investiator repeats the content analysis and compare
how do you test the consistency between the original CA and the repeats?
content analysis and coding
correlation study e.g. spearman’s rho.
0.8 or higher is accepted as reliable
what are the strengths and limitations of content analysis?
content analysis and coding
S: high external validity (generalisable), easy to gather sample
L: possible observer bias, the ‘artefacts’ are written for purposes other than the research so may lack validity
what is thematic analysis?
content analysis and coding
when researchers read the text first and then allow themes to emerge. not predetermined
how do you perform a thematic analysis?
content analysis and coding
- gather sample of texts
- read texts and spot patterns that can be coded
- re-read and look for the themes.
what are the strengths and limitations of thematic analysis?
content analysis and coding
S: stops observer bias as theories come after theme discovery, same as content analysis
L: same as content analysis
