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;
59 Cards in this Set
- Front
- Back
What do we measure in social science research?
|
behaviors
attitudes cognition's artifacts |
|
4 steps of developing indicators
|
1. define the concepts
2. determine examples of the concepts 3. compile list of indicators 4. assign #s to indicators |
|
How does science differ from other forms of knowledge?
|
science is systematic, it separates the creation of knowledge from idiosyncratic influence (individual ways of looking at things)
|
|
Requirements for scientific knowledge claims
|
-Depend on empirical verification
-"value free" (not about what is good/bad) -Derived from explicit methods so others can replicate -Are cumulative -Are falsifiable |
|
Hypothesis
|
A statement about the relationships between variables
|
|
Variables
|
things that change from unit to unit
|
|
Credibility
|
Content
Authority Critical standards |
|
Literature Review Steps
|
1. statement of problem
2. summary of previous research findings on topic 3. critical analysis 4. generate hypothesis & questions |
|
Margin of error
|
a range of values within which the population value is likely to fall
rule of thumb: 1/SQUARE ROOT n |
|
Good theories are:
|
Falsifiable
Supported by evidence General Simple |
|
Theory
|
Story of the casual relationship among concepts
(a set of systematically related statement about relationships among concepts with the purpose of explaining some phenomenon) -Aims to identify clearly & precisely |
|
Biased sampling frame
|
One in which some members of the population are more likely to be selected than others
ex: volunteer sampling |
|
Simple random sampling
|
Every unit in the sampling frame has an equal chance of being selected
|
|
3 steps of simple random sampling
|
1. Get a list of everyone in a population
2. generate the appropriate random #s 3. select individuals corresponding with your random # |
|
4 steps of systematic sampling
|
1. get a list of everyone in population
2. calculate skip interval= pop. size/sample size 3. pick a random point between 1 & the skip interval # 4. make sure every person has an equal of being selected, check to see that the list doesn't create bias |
|
Sampling frame
|
list from which units are randomly sampled to be included in the survey
|
|
Stratified random sampling
|
the population is divided into homogeneous sub-groups and simple random sampling is done within each of the sub groups
|
|
Steps for stratified random sampling
|
1. break population into relevant groups (e.g. African Americans, Latinos)
2. conduct simple random samples among each subgroup |
|
Cluster sampling
|
Used when: no lists are available, other methods are too costly
3 steps: 1. specify groups of individuals you wish to sample 2. randomly selecting some of the groups 3. survey all of the people in the selected groups |
|
Criteria to establish causality
|
-Temporal order (the IV must come before the DV)
- Covariation: the variable must be related - Non-spuriousness: on occasion an IV is implemented and a change is noted, but the change is not the result of the IV, it could be due to something else) |
|
Random assignments VS. Random selection
|
random assignments: take care of confounding variables
random selection: allows you to generalize to whole population |
|
Relationship hypothesis
|
there is a direct (positive) or indirect (negative) relationship between the IV and DV
|
|
Falsifiable claims
|
can be tested in principle
|
|
The goals of science
|
Description
Prediction Explanation Control |
|
Why is science probabilistic?
|
- relationships are not certain
- it is impossible to study all cases - relationships can change with time - relationships can vary from case to case |
|
Concept
|
An abstraction that describes a portion of reality
- cannot be observed directly - measured with variables |
|
Treatment
|
an explanatory variable that is randomly assigned to experimental units
|
|
Control group
|
group of experimental units that does not receive the treatment
|
|
Confounding variable
|
one whose effect on the response cannot be separated from the explanatory variable
|
|
Interacting variable
|
a variable that is related to differing impacts of the treatment variable
(ex: ads having different effects from Republicans & democrats) |
|
Generalizability- Ecological validity
|
Whether or not experimental conditions reflect the impact the variable has in everyday life
|
|
Field experiments VS. Lab experiments
|
are performed in as natural a setting as possible VS. conducted in a lab which may be somewhat artificial
|
|
Pairing or Blocking
|
means exposing similar (or the same) individuals to both the treatment or control
-makes estimates more accurate |
|
The main problem with observational studies
|
we cannot exclude the effects of confounding variables
|
|
Nominal variables
|
variables we can put in a category but do not have a logical ordering
(ex: gender, coded as 1 for male or 0 for female) |
|
Ordinal variable
|
categories that may have a natural ordering
(ex: 1=strong agree, 2, 3, 4, 5=strongly disagree) |
|
Ratio variable
|
A variable in which it makes sense to talk about ratios
- have a meaningful zero point |
|
Categorical variables
|
Nominal & Ordinal variables
|
|
Measurement variables
|
Ratio variables
|
|
Hawthorne effect
|
the process of being a research subject causes change in behavior
|
|
Archives VS. Observation
|
existing materials (e.g. official records) VS. watching w/o subject knowing
|
|
Likert-type scales
|
questionnaire item requiring respondents to specify their level of agreement to a statement
-meant to be summed -represents multiple indicators -must be + related -are ordinal |
|
Unintentional bias
|
questions that are easy to misinterpret
|
|
Open vs. Closed questions
|
open: hard to code, many different responses
closed: options are offered tend to be selected more than they would have been w/ open questions |
|
Criteria for evaluating variables
|
Validity
Reliability Bias |
|
Validity
|
how well a measure actually measures
|
|
Reliability
|
whether the measure gives approximately the same answer time after time
|
|
Bias
|
whether a measure is systematically off the mark in one direction
|
|
Sampling
|
A tool used to learn about a population
- selecting a sample to represent the population |
|
Population proportion VS. Sample proportion
|
the proportion of people in the population with a given characteristic
VS. the proportion of people in a given sample with a given characteristic |
|
Gallups method
Random Digit Dialing (RDD) |
What is the population of interest?
- choose a method to sample target population |
|
The biggest difficulty with polls
|
1. drawing an unbiased sample of the population
2. getting people to respond |
|
4 levels of measurement
|
nominal
ordinal interval ratio |
|
7 types of question wording effects
|
1. Deliberate bias
2. Unintentional bias 3. Desire to please 4. Asking the uninformed 5. Unnecessary complexity 6. Ordering of questions 7. Confidentiality and anonymity |
|
Discrete variable
|
one for which you can actually count the possible responses (e.g. 0,1,2,3..)
|
|
Continuous variable
|
can be anything within a given interval (e.g. age, 2 1/2)
|
|
Difficulties with observational studies
|
Confounding variables
Generalizability, extending the results inappropriately |
|
Difficulties with experiments
|
-Confounding variables: solve with randomization
-Interacting variables -Placebo, Hawthorne, experimenter effects -Ecological validity and generalization |
|
7 critical components of research (statistical) studies
|
1. What is the source of research or funding?
2. What was the nature of the contact between the researcher & the participants? 3. Who were the individuals studies & how were they selected? 4. What was the exact nature of measurements or questions asked? 5. What was the setting in which measurement were taken? 6. What are the differences in group being compared in addition to the factor of interest? 7. To what extent were the claimed effects explained? |