Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
135 Cards in this Set
- Front
- Back
contigencency tables |
unweidly for interval/ratio ( or even large ordinal) variables |
|
mutivirate associations |
1: independent 2: intervening 3: interactive 4: spurious |
|
independent relationships |
each X has an independent effect on Y |
|
interveining relationships |
all or part of an association between one variable to another is indirect - a third variable intervenes between the two |
|
interactive relationships |
one variable changes the relationship between 2 others , - effects of X1 on Y varies by category of X2 |
|
spurious relationships |
a false association between X1 and Y , caused by an antecedent variable, related to both |
|
qualitative |
inductive, interpretivists, often constructionist, use of naturalistic perspective |
|
place of qualitative theory |
grounded theory, interactive process |
|
kinds of qualitative research |
ethonography/particiapnt observation focus groups, conversation/discourse analysis, textual and document analysis |
|
main goals of qualitative research |
1: empathy 2: in depth description and emphasis on context 3: emphasis on process 4: flexability and limited structure
|
|
empathy |
seeing through the eyes of the people studied involving probing beneath the surface of social behaviour |
|
indepth description and emphasis on context |
has narrow contexts, broader social context and cultural relativism |
|
emphasis on process |
showing how events and patterns unfold over time , a long time spent in the field allows the researcher to understand the indivala and social change |
|
flexability and limited structure |
questions asked the people studied tend to be quite general , especially early in the research |
|
trustworhtiness characteristcis |
1: crediability 2: transferability 3: dependability 4: confirmability |
|
credability |
parallels measurement and esp iternal validity how believable are the findings ? |
|
types of documents |
1: personal documents 2: offical documents 3: mass media output 4: internet resources
|
|
personal documents |
letters, diaries, photos |
|
offical documents |
gov. documents and private sources |
|
mass media output |
radio, Tv programs , newspapers |
|
internet resources |
web pages, chat rooms , etc |
|
the most common method of qualitative approach is |
a case study |
|
approaches to coding |
maifest and latent |
|
manifest coding |
emphasis on what is easily quantifiable/objective - place of event , the country that the athlete represents , physical appearance, beauty , fashion etc |
|
latent coding |
semantic analyiss, emphasis is less straight forward, underlying meaning , more interpretive , portrayed as a complaint , passive, emotional , family centred. etc |
|
element or a unit |
a single case in the population , usually a person , |
|
population |
the cases you would like to be able to say something about the cases from which you will select a sample
ex. could look at women of working age if wanted to find we want to look at which aspects of women current work lives impact family size in canada |
|
sample |
a # of units selected from the population from who data will be collected
ex. might be 1000 women of working age to whom we administer a survey |
|
census |
collecting data from all units in the population, not a sample |
|
census is not a |
sample |
|
probability sampleing |
sample selected using a random process , each unit in the population has a known probability of being sampled |
|
non probability sample |
a sample selected using a non random method |
|
sampling frame |
list of all units in your population from which a probability sample will be randomly selected |
|
representative sample |
contains the same distributions of characteristics as we see in the population , we need this to generalize our findings |
|
sampling error |
errors of estimation that occurs as a result of direness between the characteristics of the sample and those of the population |
|
non - response |
occurs if an element selected for the sample does not supply the required data |
|
sampling related error |
1: not using a random method to pick the sample 2: sampling frame 3: non response
|
|
sampling frame |
isn't complete or is inaccurate, your frame does not reflect the population , so how can your sample? |
|
non repsonse |
occurs suh that those who participate in the study differ in some important way from those who do not |
|
not using a random method to pick the sample |
issuesof bias, ex. standing outside in the daytime interviewing students at western and picing who you want |
|
probability samples |
perferred in most research - most likely to be representative - can draw valid generations from our sample - based on probability/sampling theory |
|
non probability samples |
perferred in exploratory research/pilot studies - not random -less likely to be representative - cannot generalize beyond the sample -limited info |
|
simple random samples |
most basic type of prob sample - each unit has equal chance of selection ( sampling ratio works here) |
|
how to select simple random sample |
1: clearly define you population 2: devise a sampling frame for your pop 3: choose a sample size 4: # each unit of your frame from 1 - N 5: make a random # , none that are larger than N 6: each random # will correspond to a unit listen on your frame |
|
stratified random sample |
dividde population into subgroups , in our sampling frame , groups that we think are important to our study |
|
systematic sample |
- chose a pop, create a sampling frame and choose sample size , then select every 1th unit |
|
whats a potential problem for systematic sample |
periodicity |
|
periodicity |
if the cases int he sampling frame are arranged in some systematic order that corresponds to your sample interval |
|
multi stage cluster sample |
- used for sampling - large populations - populations where there is no proper sampling frame -randomly sample these |
|
trustworthiness characteristics |
1: crediability 2: transferability 3: dependabiity 4: confirmability |
|
credability |
paralllels measurement and internal validity - how believable are the findings |
|
transferability |
parallels external validity -focus on thick descriptions often look and compare rare cases |
|
dependability |
parallels reliability -long and difficult ,not popular - |
|
confirmability |
parallels replicability and objectivity - |
|
types of interviews |
1: semi structured 2: unstrucutred 3: life history 4: survey ( structured) 5: focus groups 6: particiapnt obervation |
|
semi strcutured |
have a simple, changeable interview schedule , usually called an interview guide to guide the research |
|
unstrucutred |
have little or no interview schedule |
|
life history |
usually used with personal documents - builds the story of one person or one family over the course of their lives - usually retrospective ex. sanchez family in mexico |
|
survey structured interviews |
standardization interviews sceduale - clear bengingin and end - often closed ended professsional tone interviewer is neutral |
|
focus groups |
in depth interviews 6-10 people usually meet on one occasion , usually recorded |
|
active role |
doing what the subjects are doing - sometimes can be physically dangerous ,illegal etc. |
|
going nattive |
when you genually join the subjects group you are studying ex. white streets cornor society |
|
access |
1: open setting 2: closed setting 3: covert observaton 4: overt observation |
|
open setting |
public setting , parks down town etc |
|
close setting |
private or restricted setting ( meetings of private clubs etc |
|
covert observation |
people being studied do not know they are being observed by a researcher |
|
overt observation |
people being studied know they are being observed |
|
how do you access closed settings or gain entre as a covert observer? |
- need a sponsor ( someone to vouch for you), sometimes very difficult |
|
sponsors can also serve as key informants |
person who are particularly knowledegable and cooperative |
|
drawbacks of key informants |
- researcher may ignore other group memebrs - views may not be representative of the group as a whole
- still need permission |
|
ethnogaphic data ? |
1: mental nots 2: jotted notes 3: direct observation notes 4: analytic memos 5: personal notes |
|
mental notes |
memory , when not possible to write anything down ex. during covert observation |
|
jotted notes |
written on the fly in the field , temporary reminders of whats happening as it unfolds |
|
direct observation notes |
recorded immediately after leaving the field - ordered by date, time and place for each entry |
|
analyitic memos |
notes on how to proceed, typically involves theorretical insights |
|
personal notes |
for purposes of reflexivity ( feelings and emotional reactions to reaseach etc) |
|
mode |
most occurring value , |
|
mode |
used in all types of data |
|
mean |
average |
|
median |
mid point in the distribution - |
|
median |
used with interval ratio/ordinal but not nominal |
|
nominal variables |
composed of varibles that arent linked together |
|
ordinal variables |
can be ranked , words ex level of satisfaction |
|
internal/ ratio |
# ranked ex how depressed you are from 1-10 , continuous |
|
standard deviation |
measures of variation around the mean. effected by outliers |
|
range |
measured by measuring dispersions . diff b/w the may/min variable |
|
contingency tables |
most flexible for analyizing relationships |
|
contingency tables |
2 elements are analyzed , looks for patterns of association |
|
in contingency tables % represents |
independent variable |
|
goals of quantitative research |
systematic measurements |
|
goals of quantitative research |
establishing casualty |
|
goals of quantitative research |
generalization |
|
goals of quantitative research |
replication |
|
systematic measuerment |
measurment valdiity |
|
establishing causality |
internal validiityy |
|
generalization |
external validity |
|
replication |
standardization |
|
goals of qualitative research |
empathy |
|
goals of qualitative research |
indepth description and emphasis on context |
|
goals of qualitative research |
empahsis on process |
|
goals of qualitative research |
flexability and limited structure |
|
sampling ratio |
tells us what the chance is of being selected |
|
sample weight |
tells us how many cases in the population each sample unit represents |
|
we can use the sample weight to define the |
sampling interval for systtematic sample |
|
each subgroup is sampled separately , but all use the same |
sampling ratio |
|
why is sample size very important? |
- increasing sample quality and lowering sample error |
|
as sample size increases |
error decreases |
|
temporarl order |
the cause of an event must occur before the event being explained |
|
why is temporal order relevant ? |
for establishing time order , we need a why? |
|
independent variable is the |
explanatoy variable |
|
dependent variable |
is the response variable |
|
how can we tell what is cause and what is effect? |
1: temporarl order 2: assocation 3: eliminating alternatices ( spuriousness) |
|
association between variables |
one variable ( X) changes in conjunction with another ( Y) |
|
reciprocial relationships |
direction of causation between 2 variables moves in both directions |
|
elminate ( rule out ) spurious relationships |
apparent causal relationship , |
|
eliminate spurious relationships |
ex. document an association between eating ice-cream and drowning |
|
sufficent cause |
is not necessary to produce an effect , but can produce an effect all by itself - there can also be other potential causes |
|
contribuitory cause |
is neither a necessary nor sufficient cause of Y but changes the likelihood of Y occurring |
|
univaritate |
single variable |
|
univariate |
frequency tables, graphs, measures of central tendency, measures of dispersion |
|
biviratie |
contigency tables/cross tabs , tests of association |
|
multivariate |
types of associations |
|
frequncy distributions |
simple count across categories of a single variable - mutually exhaustive and exclusive - possible with variables of all levels of measurement |
|
% frequency |
easier to understand the size difference between categories |
|
IQV |
for nominal variables
|
|
IQV |
ranges between 0 and 1 |
|
IQV |
0 if all the cases fall into a single category |
|
IQV |
1 if all the cases are equally distributed across categories |
|
IQV stands for |
index of qualitative research |
|
what is good for any ranked variable |
median |
|
what is good is the categories are large |
median |
|
the higher the standard devation |
the higher the vairablity of the mean |
|
bar charts / histograms |
good for nominal and ordinal variables and I/R if the variables are small |
|
line graph / polygon |
good for ordinal variables with lots of categories and I/R data |
|
cumulative frequency distrbutions |
% of cases in each category added to the % in proceeding categories , most useful for rank ordered variables |