ch 5 - designing a study

Question 1

soft data

Answer 1

form of impressions, words, sentences, photos, symbols

Question 2

hard data

Answer 2

numbers

Question 3

qualitative researchers

Answer 3

interpretive social science, nonlinear research path, speak a language of cases and contexts - detailed examination of cases that arise in the natural flow of social life

Question 4

quantitative researchers

Answer 4

posivitist approach, linear research path, language of variables and hypotheses, emphasize precisely measuring variable and testing hypotheses that are linked to general causal explanations

Question 5

Linear research path

Answer 5

fixed sequence of steps - staircase in one clear direction - logical step by step straight line

Question 6

Non linear research path

Answer 6

successive passes through steps, sometimes moving backward and sideways before moving on.

Spiral moving slowly upward but not directly

each cycle or repetition a researcher collects new data and gains new insights

Question 7

Selecting a topic

Answer 7

Qualitative - begin with very general research questions and the specific topic emerges slowly during the study - combine focusing on a specific questions with the process of deciding the details of study design that occurs while they are gathering data.

Quantitative researchers - narrow a topic into a focused question as a discrete planning step before they finalize the study design - then develop testable hypothesis

Question 8

Major limitations

Answer 8

-time
-costs
-access to resources
-approval by authorities
- ethical concerns
-expertise

Question 9

Qualitative Design Issues

Answer 9

The languages of cases and contexts
- examine social processes and cases in their social context - interpretations or the creation of meaning in specific settings

Critical Context
- qualitative researchers emphasize the social context for understanding the social world

Case and Process
- cases are usually the same unit as a unit of analysis or the unit on which variables are measured
- measure variables of their hypotheses across many cases
- focus on cases more than variables
- passage of time is integral

Interpretation
- assignment of significance or of a coherent meaning to something - rearranging examining, discussing in a way that conveys an authentic voice or that remains true to the original people and situations studied

Question 10

Quantitative design issues

Answer 10

Language of variables and hypotheses
- variables represent concepts
- the values or categories of a variable are its attributes

Question 11

5 characteristics of a causal hypothesis

Answer 11

1. it has at least two variables
2. it expresses a causal cause and effect relationship between the variables
3. it can be expressed as a prediction or an expected future outcome
4. it is logically linked to a research question and a theory
5. it is falsifiable- capable of being tested against empirical evidence and shown to be true or false

Question 12

Level of analysis

Answer 12

level of social reality to which theoretical explanations refer
- varies on a continuum from the micro level to the macro level

Question 13

Unit of analysis

Answer 13

type of unit a researcher uses when measuring

Common units are the individual, the group, the organization, the social category, the social institution, the society

determines how a researcher measures variables or themes

Question 14

Ecological Fallacy

Answer 14

Mismatch of units of analysis - poor fit btw units researcher has empirical evidence and the units about which they want to make statements - imprecise reasoning

Occurs when researcher gathers data at a higher or an aggregated unit of analysis but wants to make a statement about a lower or diaggregated unit

Gather data for organizations countries then draws conclusions about the behavior of individuals from those data

many low income, half live in poverty, fewer motorcycles - ecological fallacy to say rich people are more likely to own motorcycles bc we don’t know which families own motorcycles

Question 15

Reductionism/fallacy of nonequivalence

Answer 15

Another problem mismatched units of analysis and imprecise reasoning about evidence

researcher explains macro-level events but has evidence only about specific individuals

Researcher observes a lower or disaggregated unit of analysis but makes statements about the operations of higher - easier to get data on individuals so this occurs

WW1 occurred because serbian shot an arch duke - macro political event cant be reduced

Question 16

Spuriousness

Answer 16

eliminating alternatives - eliminate possible spurious factors that contribute to an apparent causal relationship -unseen 3rd factor is the real cause

calling a relationship spurious means that it is false - association might not be the basis for causal relationship - might just be illusion

Question 17

Simpson’s paradox

Answer 17

apparent peculiar situation where diffierences between groups tend to disappear or even reverse when the groups are combined

Question 18

tautology

Answer 18

refers to circular reasoning and can often be detected when the first half of a sentence appears to be a rephrasing

Question 19

teleology

Answer 19

an argument that explains that cause-effect relationship as one that fulfills a function or ultimate purpose.

Question 20

• qualitative

Answer 20

s less linear, emphasizes social context and understandings/meanings of concepts, and often focusses on
relatively few cases

Question 21

• quantitative

Answer 21

is more linear, emphasizes variables (and their attributes) and a large # of cases, and often tests hypotheses
(being aware of faulty explanations – e.g. spuriousness)

Question 22

Qualitative Design Issues:

Answer 22

• emphasizes importance of social context for understanding
  behaviour; e.g. different understandings of corruption
• focusses on (relatively few) cases, as opposed to variables

Question 23

Quantitative Design Issues:

Answer 23

variable: the empirical measure of a concept, which takes on
different values; “a variable must vary”
* attributes: the categories or levels of a variable; e.g. plumber,
banker, programmer

typically distinguish between different types of variables:
* Independent (X): the one you hypothesize has an effect on Y
* Dependent (Y): the one you hypothesize is being acted upon by X
* Intervening/3rd (Z): to test for spuriousness and develop more
complete models

• hypothesis: based on a theory and research question, it is a testable
  (falsifiable) statement about the relationship between (at least) two
  variables
• we test which of the following hypotheses is most likely to be true:
• null: there is no relationship between X and Y
• alternative (research): there is a relationship between X and Y;
  increased literacy leads to lower levels of political corruption
• we usually test more than one hypothesis