data handling and analysis

• what is primary data?

  -data collected from first hand experience

  -it is gathered for the purpose of studying specific research hypotheses
• what is a strength of primary data?

  -the specific focus of such data increases validity
• what is a weakness of primary data?

  -often time consuming and expensive to obtain and analyse
• what is secondary data?

  -data that has already been published in some form
• what is a strength of secondary data?

  -relatively inexpensive to obtain
• what is a weakness of secondary data?

  -the data that has been collected has been analysed according to someone else‘s aims and hypotheses

  -these may not match those of the current investigation
• what is meta analysis?

  -research method where rather than conducting new research, the primary data from other studies are re-analysed

  -it is used to produce an over view of existing research
• what is a strength of meta analysis?

  -it may be possible to identify common trends that aren’t apparent or not significant in single individual studies
• what is a weakness of meta analysis?

  -it is important that the criteria for including studies are applied consistently which may mean that studies need to be rejected
• what is the difference between quantitative and qualitative data?

  -quantitative is numerical and qualitative descriptive data

  -experiments gather quantitative data

  -qualitative data is gathered in self report methods such as questionnaires
• when should the median be used?

  -it is used when people express a preference

  -this is called ordinal data because it is concerned with identifying the order of people’s responses
• when should the mean be used?

  -used to represent data in which the numbers are equally spaced, this is called interval type data
• what is nominal data?

  -this refers to data that is based on categories or frequencies
• what is ordinal data?

  -allows order of preference to be established, however it cannot be assumed that categories are all equally spaced
• what is interval data?

  -based on scales which is equally spaced

  -numbers are separated by equal intervals
• percentages:

  -percent means per 100

  -percentages can be useful because they enable scores to be compared

  -percentages can also be expressed as decimal fractions
• correlations:

  -scattergrams are used to show relationships between variables

  -the scattergram is an illustration of the correlation but the correlation itself is determined by the correlation coefficient (numerical result)
• bar charts:

  -they are used when the data is in discontinuous categories

  -the bars are always of the same width and separated by a gap n
• histograms:

  -is used to represent data that is continuous

  -frequency is recorded on the vertical axis and the continuous variable is recorded on the horizontal axis
• line graphs:

  -it can give a clear view of how two or more sets of data compare
• scattergrams:

  -used to represent relationships between 2 variables

  -they are made by plotting one score on the 'y' axis and one score on the 'x' axis
• tables:

  -another way of summarising data is to use tables

  -the appropriate measures of central tendency and dispersion need to be calculated
• how can qualitative data be presented?

  -change it to frequency data by completing a tally chart

  -it is very important to be clear about how the categories are being defined and measured
• normal distribution:

  -when the mean, median and mode of a distribution are the same, you get a very particular shape

  -most scores would occur around the middle with few being clustered as they occur above and below the mean
• skewed distribution:

  -some distributions have most scores above the mean and this is called a negatively skewed distribution

  -some distributions have most scores falling below the mean and this is called a positively skewed distribution

  -they show that the mean is not a very representative score