descriptive statistics

• what are measures of central tendency?

  • mean
  • median
  • mode
• What are measures of central tendency used for?

  • reducing large data sets into averages
• Evaluation of using mean?


  • uses all data in the set which makes it accurate
  • however it can be skewed by one or two scores due to measurement error etc - this means the mean wouldn't be representative of the data set
• Evaluation of using median?

  • won't be skewed by scores/outliers
  • may not be representative of central tendency (especially if the data set is small) so it's less accurate
• Evaluations of using mode?

  • won't be skewed
  • makes more sense for presenting the central tendency in data sets with whole numbers
  • however it does not use all the data in a set
  • data set may have more than one mode
• What are measures of dispersion?

  • they quantify how much scores in a data set vary
  • range and standard deviation
• Evaluation of using range as a measure of dispersion?

  • easy and quick to calculate
  • will take into account anomalies as you are using the highest and lowest value
  • does not account for the distribution of scores
  • can be skewed by anomalies
• What is standard deviation?

  • a measure of how much numbers in a data set deviate from the mean
• How do you calculate standard deviation?

  1. calculate the mean
  2. subtract the mean from each number in the data set
  3. square those numbers
  4. add all the numbers together
  5. divide the result by the amount of numbers and square root the answer
• evaluation of standard deviation?

  • less skewed by scores
  • takes longer to calculate
• What is a normal distribution?

  • a data set that has the majority of scores near the mean average
  • data is symmetrical (There are an equal number of scores above the mean as below it)
• is IQ an example of a normal or skewed distribution?

  Normal distribution

  
• What are skewed distributions?

  • data sets that are not symmetrical
  • scores are not distributed evenly either side of the mean
  • often caused by outliers
  • can be positive or negative
• • In a positively skewed distribution the mean is much higher than most of the scores
  • in a negatively skewed distribution the mean is much lower than most of the scores
• Is this positively or negatively skewed?

  Negatively skewed

  
• Correlations

  • refer to how closely two or more things are related
  • they are measured using correlation coefficients (r)
  • if r is + 1 there is a positive correlation and if r is - 1 there is a negative correlation
• bar charts are used for discrete/quantitative data

  • e,g, loftus and palmer's study into effects of different leading questions on memory
• • a histogram is used for continuous or interval data
  • no gaps between bars
• What is the point of using inferential testing?

  • to see if a study's results are significant
  • whether observed effects are as a result of whatever is being studied rather than just random chance
• • the lower the value of p (probability) the more significant your results are
  • You can never get a p value of 0
  • In psychology, this threshold is usually < 0.05, which means there is a less than 5% chance the observed effect is due to luck
• What is a type 1 error?

  • false positive
  • when researchers conclude there is an effect but it's actually due to chance
• What is a type 2 error?

  • false negative
  • researchers conclude there is no effect but there is actually an effect and results are not due to chance
• How do you reduce the likelihood of type 1 and 2 errors?

  Increasing sample size
• What to look for when choosing which statistical test to carry out

  • whether you are looking for a difference or correlation
  • whether data is related or unrelated
  • the level of measurement
• How do you carry out a sign test

  1. work out the difference between the two sets of data
  2. add the total numbers of + and - (where there is no difference data can be ignored)
  3. the least frequent sign is your S value
  4. the value of n is the number of participants (not including the participants who had no difference in data)
  5. compare calculated value to critical value
• What is a 2-tailed test

  • non-directional hypothesis
• What is a one-tailed test

  • directional hypothesis
• What is nominal data?

  • data is categorised
  • variables are discrete and don't overlap
  • categories have no natural order
• What is ordinal data?

  • data that is put in an order or rank
  • but the difference between each point is not consistent
  • e.g. First, second, and third place in a race or Ranking your mood on a scale of 1-10
• What is nominal data?

  Data which can be separated into discrete categories which do not overlap
• What is interval data?

  • A measurement taken on scale and the unit is equally sized and objective (e.g temperature in Celsius or Fahrenheit)
• What is ratio data?

  • interval data
  • must be numerical values where the difference between points is standardised
  • not possible to have negative values 
• Discrete data - information that can only take certain values e.g. number of clicks on social media

  Continuous data - numerical data that can be measured in units, e.g. height