Levels of measurement

• What do level of measurements tell us


  tell us how the DV was measured + therefore how precise the data is
• What is nominal data


  • weakest level of measurement:

  -data is in named (behavioural) categories

  -shows the frequency/number of times something occurred e.g. the number of people who passed their driving test

  -likely to be collected from closed Qs in self reports or from structured observations

  (simplest form of data)
• What is ordinal data


  • more precise than nominal data

  -data is placed in rank order (e.g positions such as 1st, 2nd, 3rd or highest to lowest or happiness score on a scale of 1-10)

  -the difference between each rating, rank or score is not known + doesn't have to be equal

  -can be calculated from a rating scale (e.g likert/ranking scale) question or test score
• What is interval data


  • most precise level of data

  -data is on an objective scale of known units with regular/equal intervals e.g time, length, temp
• Levels of measurement examples

  • running speed:

  -nominal = fast or slow

  -ordinal = 1st, 2nd, 3rd in a race

  -interval = time taken in seconds to run 100m

  • height:

  -nominal = short or tall:

  -ordinal = tallest to shortest

  -interval = measurement in feet + inches

  • performance in a psychology test:

  -pass or fail

  -highest to lowest

  -percentage score
• What level of measurement can the mode be used for


  -can be used for any (nominal, ordinal + interval)

  -mostly used for nominal data as it may not always be numerical in form so the mode is the only appropriate measure to use
• What level of measurement can the median be used for


  -ordinal + interval

  -can only be used on data that is or can be ordered
• What level of measurement can the mean be used for


  -interval

  -the mean requires equal spacing between adjacent values or scores in the scale
• Strength + weakness of nominal data


  +easy to generate from closed Qs; large amounts of responses can be collected quickly = increasing external reliability

  -leat precise: w/o linear scale Ps may be unable to express degrees of responses

  -can only use the mode as a MoCT
• Strength + weakness of ordinal data


  +indicates relative values on a linear scale instead of just totals so is more informative than nominal data

  -we don't know how large the gap is between the 1st and 2nd place or between 2nd and 3rd place or if these gaps are equal
• Strengths of interval data


  +most precise: more informative than ordinal and nominal as the points are directly comparable because they are all of equal value

  +scientific measures used to record the distance between values are highly reliable