Data handling

• Primary data


  Original data collected specifically towards a research aims
• Primary data examples


  Researchers interested in the effects of machines induced workplace stress can gather primary data specifically related to the issue
• Secondary data


  Data originally collected towards another research aim which has been published before
• Secondary data examples


  Government and public sector reports, websites and books
• Primary data advantages


  ✅researcher has control over data as collection is designed to fit aims and hypothesis of the study

  ✅more reliable and valid
• Primary data disadvantages


  ❌time consuming and expensive to obtain and analyse
• Secondary data advantages


  ✅inexpensive

  ✅readily available

  ✅when drawn from several sources it can give clearer insight to research

  ✅data may have already been subjected to statistical testing and is known if it's significant
• Secondary data disadvantages


  ❌may have inherent biases as it's undergone some sort of interpretation

  ❌for some studies data may not fit the needs of the study
• meta-analysis


  a procedure for combining the results of many different research studies into a larger study to allow identification of trends and relationships
• Meta-analysis Advantages


  ✅helpful when studies have found contradictory or weak results and helps gives a clearer view of the overall picture
• Meta-analysis disadvantages


  ❌criteria for including studies is very strict

  ❌relies on primary research being good quality as it uses secondary data
• measures of central tendency

  mean, median, mode
• measures of dispersion


  range and standard deviation
• Mean


  Worked out by adding up all values and dividing by total numbers of values. It's the best measure of central tendency only if there aren't outliers

  can't be used with ordinal data
• Median


  Middle value in a set of numbers that have been out in order and is better than mean if there are outliers

  Can be used with ordinal data

  Not good with small data sets as it may be a poor representation of the middle
• Mode


  Most frequently occurring value in a data set

  Must use with nominal data

  Unaffected by outliers

  Not necessarily the middle of the data
• Range


  Working out difference between highest and lowest score in data. The higher the number the higher the spread in the data

  Badly affected by outliers
• skewed distribution


  Median is best measure of central tendency

  Standard deviation is best measure of dispersion
• negative skewed distribution


  Mean < median

  Outliers are at the bottom of distribution
• Positive skewed distribution


  mean > median

  Outliers are at the top of the distribution
• normal distribution 3 standard deviations

  -1SD to 1 SD - 68% so 34% each way

  -2SD to 2SD - 95% so 14% each way

  -3SD to 3SD - 99% so 2% each way
• How to calculate standard deviation

  1. Calculate mean

  2. Square root the sum of each individual score minus the mean put in brackets and square it, all over the amount of scores -1
• Nominal Data


  data of categories only. Data cannot be arranged in an ordering scheme. (Gender, Race, Religion)
• ordinal data


  data exists in categories that are ordered but differences cannot be determined or they are meaningless. (Example: 1st, 2nd, 3rd)
• Interval Data


  Differences between values can be found, but there is no absolute 0. (Temp. and Time)