L9-15 research methods Flashcards
(31 cards)
what are the three levels of measurement
- ordinal
- nominal
- interval
what is ordinal data and what central tendancy measure/s uses it
a numerical scale that can be ordered but the units are not of an equal precise size
median and range
what is nominal data and what measure of central tendancy uses it
data seperated into a set number of categories
mode
what is interval data and what measure of central tendancy does it use
standardised scale with equal, precisely defined scales
mean and standard deviation
what are tables used for
for ‘raw data’ which are measurements collected in a research study - numbers before any descriptive statistics have been carried out
when are barcharts used and what are their characteristics
mainly for nominal data
- there should be gaps between bars to show te lack of continuity
what are scattergrams used for
correlations (relatioships)
what are histograms used for and their features
continuous data
- bars touch each other
- the y axis represents the requency within each interval
what are line graphs used for
conitinous data
features of normal distribution graph
classic bell shaped curve
mean, median ad mode are all in the extact mid-point
distribution is symmetrical around the mid-point
dispersion of scores either side of mid-point is consistent and can be epressed in standard deviations
positively skewed disribution features
most of the scroes are bunched towards the left
mode - median - mean
negatively skewed disributions
when most of the scores are bunched towards the right
mean is always lower than mode
mean- median - mode
what is inferential statistics
the formal purpose of statistical testing (IS) is to determine the likelihood that the effect/difference/relationship/association found in the study is due to chance
this increases the scientific credibility and objectivity of research and allows the researcher to determine whether the null hypothesis should be accepted or rejected
what is the usual significance level
p<0.05
meaning the probability that the observed effect occured by chance is less than 5%
what is the difference between a one-tailed and a two-tailed test
one tailed test is a directional hypothesis
two tailed test is a non-directional hypothesis
when is the sign test used
- researchers are looking for a difference between their conditions
- a related design (repeated measures or matched pairs) has been used
- the level of measurement in the study is nominal data
how to carry out a sign test steps
- identify categories
- calculate number of participants in each category
- asssign the category when there is no difference a 0 sign, these participants are then removed from the rest of the test
- assign one of the other two cateogries a + sign and the other a - sign
- out of the + and - sign cateogries, identify the category with the smallest number. This number is S.
- S is our calculated number
what three steps are used to determine which statistical test is needed
- level of measurement
- is the study a test of difference, correlation or association
- experimental design
what are variables when testing for correlation relationship
both variables are ordinal or interval
what variables when testing for association relationship
where both variables are nominal
what is a parametric test
more poweful and robust than other tests.
- data must be at an interval level
- participants must be drawn from a normall distributed sample within the population
- there must be homogeneity of variance meaning standard deviations of conditions must be similar
how to check for a type 1 error
- keep the tailed test and N the same
- change the significant level to the smallest one you can –> go as far to the right of the critical values table as you can
- chek if the calculated value is still significant or not
- if it is you can be confident you haven’t made a type 1 error. It it’s not, you likely have made a type 1 error
what is a type 1 error
when the null hypothesis is rejcted and the alternative hypothesis is accepted when the null hypothesis is true i.e. saying a result is significant when it is due to chance.
often called error of optimists. occurs when the significants level is too lenient
what is a type 2 error
this is when the null hypothesis is accepted but should have been rejected because the alternative hypothesis is ‘true’ .
Often called the error of pessimists. Occurs when the signifinance level is too stringent.
