research methods Flashcards
One-Tailed Hypothesis/directional hypothesis
You make a prediction about the direction of your effect
For a two-tailed hypothesis
you predict A difference But don’t say what the direction of the difference will be
what is nominal data?
categorical
what is ordinal data?
everything has a category and this can be ranked but there is not distinct measurement/scale
what is interval data?
measured along a numerical scale that has equal distances between adjacent values.
what is a between-subject design?
participant experiences one condition
what is a within-subject design?
all participants experience both conditions
what test do we use to work out whether something is normally distributed?
shapiro-wilk
what must all data be before we conducted a significance test test?
normally distributed
data is significant when the p value is
p < 0.05
if something is normally distributed p is
p > 0.05
what does the p value represent?
the p-value is the probability of obtaining results or at least as extreme as the observed results of a statistical hypothesis test, assuming that the null hypothesis is correct.
what is a null hypothesis?
states that there is no difference in what you are researching
describe parametric and non parametric tests
parametric tests are more powerful and what we conduct first but they have lots of assumptions. if our data doesn’t meet these assumptions there’s usually a non-parametric alternative.
between-subject design and when you are comparing your data to one single number
one sample t-test
one sample t-test assumptions
assumptions, data are independent, continuous (interval) and the data is normally distributed.
what is the non-parametric test for one sample t?
if it doesn’t meet these assumptions then it is a one-sample Wilcoxon test, this is a non parametric test
between-subject design when comparing two conditions
independent samples T-test
what are the assumptions for independent samples T-test?
assumptions, data are independent, continuous, n=12<, data is normally distributed and homogeneity of variance.
what is homogeneity of variance?
is the variance of both conditions similar
how do we check for homogeneity of variance?
conduct a Levines test. homogenous data roughly falls on a straight line. this test shows wether there is a significant difference. we dont want there to be so p>0.05
what is the non-parametric for independent sample t test?
Mann-Whitney U test
within-subjects and matched pairs design
paired sample test
assumptions of paired sample test
assumptions, data are interval/continuous, n=12<, and the DIFFERENCES are normally distributed.
what does a paired sample test asses
asses the probability of getting a mean difference as large as you found by chance
how do we find if the differences between the two conditions is normally distributed?
conduct a Shapiro-Wilk, should be non-significant. p>0.05
what is the non parametric test for paired sample test?
Wilcoxon signed ranks test
when you have nominal data, one variable and 2 categories
inomial, one variable with two outcomes that are mutually exclusive, random sample, independent, you know the expected distribution of scores
when you have nominal data, one variable and 3+ categories
3+ options =chi-square goodness of fit
when you have nominal data, two variables
chi-square test for independence
what are the assumptions of a chi-square test for independence?
two dichotomies, male/female AND yes/no, random sample, independent, sample =40<, N=5<
what test to conduct if chi-square test for independence assumptions are not met?
fisher exact test
what tests when you have nominal data
binomial
chi-square goodness of fit
chi-square test for independence
fisher exact
what is the control variable?
things you intentionally keep the same
what are cofound variables?
effect results
influence the dv
extraneous variable
variables not controlled in the experiment
what are descriptive stats?
numbers that summarise data
e.g., mean mode median
what is a z score?
a standardized score
compares scores between participants
or across conditions
when is it best to use median?
if data is skewed
what letter represents correlation stat
r
range of correlation coefficient
-1 –> 1
1 being strong positive
-1 being strong negative
what if correlation is non linear
conduct non parametric test
non parametric for correlation
spearmens rho
kendalls t
what is a conversation analysis?
- detailed transcript of convos
- determine the means and rescources people use
example of conversation analysis being used?
comaring verb talk and speak in police negotiate situations
what is a discourse analysis?
all types of spoke interaction and written text
language as a form of action
language varying in its function
discourse analysis key terms
DA - Construction
DA - function
DA - variation
DA - function and variation
DA - accountability
DA - discursive devices
what is constructcion?
how we construct our status and ourselves
as well as other ideas and people
function
language for a purpose
variation
language varies in accordance with its function
discursive devices
Disclaimers* Provision of detail* Reported speech* Category entitlement* Extreme case formulations* Three-part lists
gibsnos analysis of milgrams studies
transcribed audio
discoursed analytical methods
considered how a substantial minority disobeyed
occasional deviations from script
most prods resembling orders appear to have been resisted
rather than being about obedience its about persuasion
what is the difference between discourse analysis and conversation analysis?
Conversation analysis is a specific type of discourse analysis that focuses on the structure and organization of spoken interaction. Discourse analysis, on the other hand, is a broader field that examines language use in a range of contexts, including written texts, speeches, and media.