Reading...
Play button
Play button
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;
33 Cards in this Set
- Front
- Back
|
ecological fallacy
|
Erroneously drawing conclusions about individuals based solely on the observation of groups
Example: study on how wealthy countries are. Decide US is wealthy. Meet american-->assume he is wealthy? NO. |
|
|
Reductionism
|
Limiting the variables under study
• Example: team loses, does that mean team stinks? |
|
|
three purposes of research design
|
a. Exploration
• Why do exploratory research? To find out if something is going on • Theory is not important. • Examples Are people migrating to california? Are? • Basic research, will not generalize or make a particular theory b. Description • Describes what is happening in the world What is happening? Who is doing it? Whats, wheres, when, how many Example • Census c. Explanation • Asks WHY • Most important research |
|
|
the 4 units of analysis
|
a. Individuals
• Most basic unit of social research b. Groups • Interested how these groups behave as a collective • Characteristics may vary, but we care about the collective of the group c. Organization • Similar to groups, but to a higher level Think of organized crime-includes: • Large gangs (group) • Small gangs (group) d. Social artifacts • Things we create Movies Sports Exams Fist fights |
|
|
Cross sectional studies
|
a study based on an observation representing a single point in time (snapshot)\
How do you feel about Barak Obama RIGHT NOW? Static, does not take into account other points in time, only right now How to get around deficiency? Longitudinal study |
|
|
Longitudinal study
|
collecting data at several points of time.
|
|
|
Types of Longitudinal Study
|
1. Trend study
2. Cohort study 3. Panel study |
|
|
Trend study
|
given characteristic of a population is monitored over time
• Same phenomenon • Studying different groups of the population |
|
|
Cohort study
|
study where some specific sub population (cohort) is studied overtime
• Same phenomenon • Same "types" of people |
|
|
Panel study
|
longitudinal study where data is collected from the same set (panel) of people
• Same phenomenon • Same people • Drawbacks: • attrition-drop out • Testing can influence • Expensive |
|
|
different steps in research design
|
a. Getting started
i. Pick a topic ii. Literature review 1. What other people found 2. How 3. Voids or what people haven't found b. Conceptualization • What do you mean by 'x' • "for purposes of study, juvenile is…and delinquency is…" c. Choice of research method • Surveys • Experiment • Qualitative • Ethnography d. Operationalization- • How are you going to answer your questions? e. Population and sampling i. Who are you going to study? ii. Determine sample and population 1. Population-the group about whom you are going to draw conclusion 2. Sample-group of people who you will actually study f. Observations g. Data processing h. Analysis • Did your hypothesis pan out? Did you reject or accept the null hypothesis? i. Application--write it up |
|
|
Conceptualization
|
process through which the meaning of a research term is specified.
• What do we agree? |
|
|
Dimension
|
specific aspect of a concept; different grouping of indicators
|
|
|
Indicator
|
a sign of the presence or absence of a concept
|
|
|
Operationalization
|
the development of specific research procedures (operations) that will result in empirical observations representing those concepts in the real world.
• Coming up w/ the questions you are going to ask to measure the concept. |
|
|
Reliability
|
Quality of measurement method that suggests that the same data would have been collected in repeated observations of the same phenomenon
|
|
|
Validity
|
The extent to which a measure actually reflects the meaning of the concept you are observing.
|
|
|
conceptualize and operationalize a concept
|
Is a restaurant good? (group)
• Define good restaurant: o A good restaurant is somewhere you would recommend your friends to. o An enjoyable experience 1. good Food (dimension) a. Indicators: i. Fresh ii. Variety 1. Measures: a. Multiple choices b. Vegetarian choices c. Drinks d. Diet options iii. ingredients iv. portions 2. Service (dimension) a. Fast (indicator) b. Friendly(indicator) i. Measures: 1. Do they have a smile? Friendly? 2. Greeting? 3. Attendance? c. Accurate (indicator) 3. Environment (dimension) a. Cleanliness (indicator) b. Appropriate atmosphere (indicator) c. Ambiance (indicator) |
|
|
different levels of measurement/How do you know if a variable is reliable and/or validity?
|
1. Reliability? reliable measure?
1. Test retest method a. Example, stepping off and on the scale, same weight 2. Split-half method a. Example: do you love barack obama? Yn? Do you hate obama? Yn? i. If yes, then no. 3. Use established measure 2. validity? create valid measures? 1. Face validity • Doesn't make sense 2. Criterion validity • The extent to which a measure relates to exterior criteria • Example: measure wealth: 3. Construct validity • The extent to which one part of a measure matches other parts of a measure • Example: Love: high emotional, high physical, should be high money love 4. Content Validity • Whether or not a measure covers all the elements or detentions of a concept • Example: measuring depression • Do you wear black? Yn • Do you eat loads of noodles? Yn • Do you like the smiths? Yn • Do you sleep a lot? Yn • Not content validity 5. Use of established methods • Used for years and years\ 6. Tension between reliability and validity |
|
|
Index
|
accumulation of scores assigned to individual attributes--same value of each number
• Example: want to buy a car. What do you consider? (right side 1/left side 0) add scores of each car.. • Prices • High or low • Resale value • Make/model • What you want/what you don't want • Mpg • Good or bad |
|
|
Scale
|
assign scores to pattern of responses (different scores based on values of scores)
• Example: high price/good mpg--some variables worth more than others |
|
|
How are indexes and scales similar and different?
|
i. Similarities
1. Both are ordinal measures (order but no distance) 2. Both are composite measure of variables ii. Differences 1. Scoring/value a. Index-accumulation of scores assigned to individual attributes--same value of each number b. Scoring- assign scores to pattern of responses (different scores based on values of scores) |
|
|
the steps in constructing an index
|
1. item selection
2. examination of empirical relationships 3. index scoring 4. index validation |
|
|
Questionnaire
|
a document containing questions and other types of items designed to solicit information appropriate for analysis (the instrument used to survey)
|
|
|
contingency questions
|
a survey question intended for only some respondents, determined by their responses to some other questions
|
|
|
response rates
|
The number of people participating in a survey divided by the number selected in the sample in the form of a percentage
• Why important? Calculate sampling error |
|
|
differences between open-ended and closed-ended questions. Advantages/disadvantages
|
• Open ended-not fixed responses
• Very valid but not necessarily reliable • Close ended-fixed responses • Reliable but not necessarily valid • People not always able to express themselves |
|
|
guidelines for asking questions/creating a questionnaire.
|
1. Chose appropriate question forms
Demographic vs. substantative questions • Demographic- • Where you are from • Age • Income • Race/ethnicity • Gender • Marital status • Substantative • Questions about subject studying • Ie, do you like barack obama? y/n Question vs. statement • Question-used to illicit a direct response to a topic • Ie: is obama a good president? y/n • Statement-Obama is a good president. y/n Open ended vs. close ended • Open ended-not fixed responses • Very valid but not necessarily reliable • Close ended-fixed responses • Reliable but not necessarily valid • People not always able to express themselves 2. Make items clear Researcher will know more on topic, so clear and simple 3. Avoid double barreled questions (those that address multiple points) 4. Respondents must be competent in answer (cant ask what age did you start talking) 5. Respondents must be willing to answer 1. Questions that matters to them 2. Get them engaged 3. Increases people's willingness and competency to answer 6. Questions should be relevant 7. Short items are best 8. Avoid negative items 9. Avoid biased terms |
|
|
Questionnaire vs survey
|
o Questionnaire-a document containing questions and other types of items designed to solicit information appropriate for analysis (the instrument used to survey)
o Survey-(verb) a process by which to have people answer questions |
|
|
process of mail distribution and return of mail surveys
|
1. Process
Make as simple as possible Card informing of coming survey Survey with cover letter (about 3-4 days later) Follow up letter (3-4 days later) Second survey w/ new letter (7-10 days later) Follow up letter Continue as able/necessary 2. Return envelopes-(prepaid envelopes) 3. Monitoring returns/follow up mailings 1. Why do it? For response rate 4. Response rate The number of people participating in a survey divided by the number selected in the sample in the form of a percentage Why important? Calculate sampling error Acceptable response rate for a mail survey • The more the better • 50% is adequate • 60% is good • 70% very good How would you increase response rate? • Incentive • $ (but can be expensive) • Raffle • Personal appeal |
|
|
What are response rates used for?
|
The number of people participating in a survey divided by the number selected in the sample in the form of a percentage
Why important? Calculate sampling error |
|
|
acceptable response rate for a mail survey
|
• The more the better
• 50% is adequate • 60% is good • 70% very good |
|
|
strengths and weaknesses of survey research
|
• Strengths
o Useful in describing the characteristics of a large population o Can do large samples o Flexible Can ask many questions on any topic Can be flexible in analysis o Measurement o Strong on reliability • Weaknesses o Does not assess attitudes, orientations, circumstances and experiences o Does not deal w/ social life o Inflexible Cannot change initial study o Artificiality-cannot measure social action Topic may not be amendable to measurement thru questionnaires The act of studying the topic-such as an attitude-may affect it o Weak on validity |
