RESARCH METHODS

avatar
Created by
lanre

Cards (17)

  • INTERFERNETIAL STATS USE A random sample of data taken form a population to help us make inferences about the population
  • Null hypothesis (H0)■ = No difference■ Alternative hypothesis (H1)■ = Difference
  • descriptive stats tell us about our sample
  • The general rule is that if your p value isbelow .05 then your result is ‘statisticallysignificant
  • P-valuesProbability the result we found (or one more extreme)occurred by chance assuming the null hypothesis is true
  • A statistically significant result is unlikely to be due to chance. It means that the result likely reflects a real effect or relationship
  • P-valuesProbability the result we found (or one more extreme)occurred by chance assuming the null hypothesis is tru
  • hus, p < .05 indicates that we should reject the null (H0)
  • A one-tailed test looks for a change in a parameter in one direction, while a two-tailed test looks for a change in both directions
  • 1-tailed tests can be used for directional hypotheses (dogs are larger than cats); the 5% significance level isconcentrated on one tail
  • -tailed tests are used for non-directional hypotheses* (dogs and cats are different sizes); the 5%significance level is split across both tails
  • or Cohen’s d, this is 0.2, 0.5, and 0.8• d tells us the size of the difference, while p tells us how surprising that difference is – but this isheavily dependent on sample size!
  • A Chi-square test measures the discrepancy between observed frequencies andexpected frequencies.• If the observed frequencies differ from expected frequencies then this suggests thatthere is an association between the IV and the DV
  • If we’d given the children an extra choice (e.g., some crisps) then we would have3 columns and 2 rows so this would be a 3x2 chi-square
  • nominal data is for chi sqaured
  • Sample size is important we need to have an appropriate number of data points. But also if thesample is too large this can lead to misleading results.
  • Smaller p-values = less likely the results are due to random chance / more likely effect is reaL