research methods Flashcards
what are inferential stats
whether changes we see are real or just due to chance
if there is a real effect we can accept the alternative hypothesis
an inferential test of significance will indicate whether we should retain or reject a hypothesis
what is chance equal to or less than
and if it is what does it mean
we accept the alternative hypothesis
0.05
1/20
what is a type 1 error
occur when the level or significance is too lenient
accepting the alternative when the results are not significant are known as a TYPE 1 ERROR
what are levels of significance
if the result or equal to less than 0.01 it may still be reported and followed up with more evidence
at 0.01 there is a chance that the alternative was accepted and shouldn’t have been
accepting the alternative when the results are not significant are known as a TYPE 1 ERROR
if the result it equal to or less than 0.01 it is highly significant and therefore not likely to be due to chance
if this was standard measure of significance we would be rejecting a lot of alternative hypothesis when there was a real effect
retaining the null when there was a real effect is known as a TYPE 2 ERROR
occurs because it is set too stringent
what is a type 2 error
occurs because it is set too stringent
retaining the null when there was a real effect is known as a TYPE 1 ERROR
what is nominal data
least precise can be used for frequency of occurrence in each category. just counting most basic form of data collection. collecting info about people.
what is ordinal data
data which has been collected is put into some type of order whether it is numerical or alphabetical rank. info comes from scores such as grades for a test or rating scales such as attractiveness 1-10
what is interval/ratio data
known differences between each data set. ratio data will have an absolute zero. measurements such as height,weight and spend are ratio because they start withzero
what are experimenter effects
the way an experimenter may influence the outcome of a experiment by their actions or mere presence
- the hawthorn effect = mere presence of a researcher can have an affect
- demand characteristics = effect of experimenter causes the participant to alter their behaviour to meet the perceived expectations of the experimenter
how can experimenter effects be controlled
standardisation = all experience the same instructions
single and double blind experiments
what is objectivity
being impartial and judgement free
important that the DV is measured objectively
what is internal validity
how well the procedure of a study established a casual relationship between the IV and DV. can be improved by having a standardised procedure. controlling order effects and individual difference whilst avoiding demand characteristics
what is external validity
the extent to which the findings apply to other people and situations
what is the mode
give advantages and disadvantages
most common score
might not represent data as most common score could be the lowest
if there are more than two modes it becomes a meaningless measure of central tendency
not affected by extreme data
what is the median
give advantages and disadvantages
can show where the majority lies
not affected by extreme scores or a skewed distribution
may not represent typical score if used on a smaller number of values
does not represent outliers
what is the mean
outliers ca skew data
need to be reasonably spread
what data does Spearman’s test apply to
correlational and ordinal data
what data does the chi squared test apply to
experimental design with an independent groups measure and nominal data
what data does the mann whitney test apply to
experimental method, independent groups design and ordinal or interval/ratio data
what data does the wilcoxon signed rank test apply to
data is ordinal or interval/ration with an experimental method and a repeated measures or matched pairs design
what is a one tailed hypothesis
direction of difference can be predicted
and directional hypothesis is stated
what is a two tailed hypothesis
direction of difference cannot be predicted and a non directional hypothesis is stated
what do mann whitney and wilcoxon require for results to be significant
the observed value to be equal or less than the critical value
what does the chi squared and spearman test require for results to be significant
the result to be equal to or greater than the critical value
what are measures of dispersion
measures of spread
how spread out data is
what is the range
the difference between the lowest and the highest
high range = spread out low range = closer together
disadvantages of the range
affected by extreme scores so may not be useful
what can you do if there are extreme scorea
calculate the interquartile range
cut out lowest and highest and calculate the range of remaining
what does the standard deviation tell us
the distance of each score from the mean
a high sd means big spread
a low sd means small spread
normal distribution is shown when the curve is..
bell shaped
68% within one SD
95% within two SD
