CHAPTER 4: RESEARCH DESIGN/LOGIC OF CAUSATION

Individualistic Fallacy

(Attributing something to a group based on the observed behaviour or characteristics of individuals. Example: Estimating the percentage of individuals in a country who favour democratic government and taking this as an indicator of the degree that the political system in that country is democratic.)

3 criteria of nomothetic causality

1) variables must be correlated- must be an empirical relationship between variables 2) cause takes place before effect- independent variable precedes effects of dependent variable 3) variables are non-spurious- the empirical correlation between two variables cannot be explained in terms of some third variable

Spurious relationship

A coincidental statistical correlation between 2 variables that is shown to be caused by some 3rd variable; when a third variable is actually doing the of the change in the dependent variable explaining (Ex. People with long hair do better in class- not to do with hair length; 3rd variable of gender (girls have longer hair) that explains why people with long hair do better; can eliminate IV; they are not responsible for the change

Intervening relationship

A relationship in which the control variable intervenes between the independent and dependent variables; the independent variable does have an effect (Drinking (IV) leads to pregnancy (DV); drinking is not DIRECTLY responsible; drinking leads to unprotected sex which leads to pregnancy); when there is a real relationship between IV and DV, but it is not a direct relationship between IV and DV; IV causes something else that causes DV; A=B and B=C; A and C are related but not directly; they are related through B

cohort study

A study in which some specific subpopulation, or cohort, is studied over time, although data may be collected from different members in each set of observations; Subjects who experienced a particular event in common (e.g., those born in the same week; graduated in the same year). An example would be a series of national surveys, conducted perhaps every 10 or 20 years, to study the economic attitudes of the cohort born during the Great Depression (1939-1945).

trend study

A type of longitudinal study in which a given characteristic of some population is monitored over time. Independent samples (i.e., not the same individuals); Example: The comparison of several censuses. Research example of some researchers who wanted to know whether drug use among Ontario youth was declining

panel study

A type of longitudinal study in which data are collected from the same set of people (the sample or panel) at several points in time. Non-independent samples (same set of people each time); By studying the same people we can better assess the effect of specific events on them, as we can compare their individual attitudes before and after. Problems: PANEL ATTRITION: Individuals dropping out by choice or due to circumstances People may become overly conscious about being observed (PANEL CONDITIONING)

grounded theory

A type of qualitative research that examines the process of a phenomenon and culminates in the generation of a theory

Correlation (first criteria for nomothetic causation)

An empirical (actual observed) relationship between 2 variables such that a) changes in one are associated with changes in the other or b) particular attributes of one variable are associated with particular attributes of the other. Correlation in and of itself does not constitute a causal relationship between 2 variables, but it is one criterion of causality

Retrospective Data

Answers to questions about experiences or events that happened a long time ago yield information. Problems: People have faulty memories and sometimes they lie, creating a very positive story from memories about early life.

Explanatory Research

Answers to why certain events happen Why people act in certain ways. Reports on the relationships among different aspects of the phenomenon under study. Causal argument. Examples: (1) Why some cities have higher crime rates than others (2) Why a pro life demonstration ended in violence; The explanation (X- Independent variable) or the "Why" of a social phenomenon (Y- dependent variable)

Social Artifacts

Any material left as the result of communication processes, in a fixed and observable state

Groups (Units of Analysis)

Can be units of analysis when we study the characteristics that belong to a group. We can often derive the characteristics of social groups from those of their individual members. (Examples: Black couples, rich families); One can define a family, for example, using it as a unit of analysis and potentially find: Total income, percentage of families whose heads have a university degree, type of dwelling, or number of members (Other groups: Married couples, friendship groups, a university class, a city, a town, or street gangs)

Nonspuriousness

Despite two variables being correlated and one preceding the other, it may be possible that neither is a cause of the other. One might make a claim that students with shoe size and math ability are correlated and that shoe size affects math ability. On investigation, one finds that it is not shoe size but age. In a spurious relationship, the independent variable has no effect on the dependent variable and can be eliminated entirely as it is a third hidden variable that is causing the effect.

3 Purposes of Research

Exploration, Description, Explanation

cross-sectional study

Observation of a population, event, or phenomenon at one point in time; A snapshot; Example: 2001 Census, 2008 General Social Survey; Explanatory and descriptive studies are often cross-sectional. PROBLEMS: (1) Difficult to generalize from observations at one point in time; (2) many explanatory studies have one-time conclusions but aim at causal processes that occur over time (see notes).

longitudinal study

Observations of the same phenomenon over an extended period of time. Observed as least twice (two waves). Types: Cohort, Panel, Trend; PROS: Accounts for change over time CONS: Costly, complex design, people may drop out (panel studies)

Individuals (Units of Analysis)

Often characterized in terms of membership to social groupings (Examples: race, class, gender, etc.); Most common unit of analysis; Applicable to all sorts of persons (Examples of persons: University students, Autoworkers, Canadian voters, Single parents)

Organizations (Units of Analysis)

Organizations (mainly a type of group) (Examples: corporations, church congregations, universities); Organizations might be characterized in terms of: Number of employees, net annual profits, gross assets, number of contracts, or in the case of a church, the belief system, number of clerics, number of members, proportion of those who are active or inactive, and the proportion of women who work full-time in the church organization

Social Artifacts (Unit of Analysis)

Products of social beings and their interactions Can range from concrete objects (books, newspapers, poems, paintings, automobiles, buildings, songs pottery, and scientific innovations) to social interactions (weddings, friendships, riots, highjackings)

Exploratory Research/Studies

Pursued through the use of focus groups or small-group discussions. Studies done for 3 purposes: 1) to SATISFY the researcher's curiosity and desire for better understanding 2) to TEST THE FEASIBILITY of undertaking a more extensive study and 3) to DEVELOP THE METHODS to be employed in any subsequent study; Often done when examining a new interest or subject- valuable for breaking new ground in an area/suggesting further research. LIMITATIONS: 1) Seldom provide satisfactory answers to research questions, though they can hint at the answers and give insights into which research methods could provide definite answers 2) Subjects often not representative of the larger population of interest

Descriptive Research

Reports on the characteristics of some population, situation, or event. Answers the questions of what, when, where, and how. No analysis or explanation of the phenomenon described. Common in ethnographic research commonly used by anthropologists. The Canadian census is an excellent example of descriptive social research. The goal of the census is to describe accurately and precisely a wide variety of characteristics of the Canadian population, as well as the populations of smaller areas such as provinces and cities.

Necessary Cause

Represents a condition that must be present for the effect to follow (Ex. it is necessary for you to take university courses in order to get a degree- without the courses, the degree never happens BUT simply taking the courses is not a sufficient cause of getting a degree- you need to take the right ones and pass them); Necessary but not enough for the effect to happen. Does not guarantee the occurrence of the effect.

Sufficient Cause

Represents a condition that, if it is present, guarantees the effect in question. Enough to achieve an outcome, although it is not the only possible cause. This does not mean that a sufficient cause is the ONLY possible cause of a particular effect (ex. Skipping an exam in this course would be a sufficient cause for failing it, though students could fail in other ways as well) - a cause can be SUFFICIENT but not NECESSARY

Panel Attrition

Some of the respondents studied in the first wave of the survey may not participate in later waves , whether by choice or by circumstance - may distort the results of a longitudinal panel study.

The cause (dependent) takes place before the effect (dependent) (second criteria for nomothetic causation)

This is the time order. Cause must precede the effect. One variable(s) (independent) must precede another (dependent). Problems: Not all causes are easy to detect before the effect: E.g. Are religious attitudes dependent on political attitudes or are political attitudes dependent on religious attitudes?

Comparing the 3 types of longitudinal studies

Trend: This will look at shifts in attitudes of the general Canadian population using polling data collected on a regular basis. Cohort: This will look at shifts in attitudes among the "World War II generation." One would study a sample of people 30-40 years old in 1952 and a new sample of people aged 40-50 in 1962. Trend: This will look at shifts in attitudes of the specific people who are surveyed over time.

Unit of Observation

What is directly observed—where the data are collected from; Can be the same as the unit of analysis, but not necessarily; What we observe directly. Sometimes our units of analysis are the same as the units of observation but not always- sometimes we study our unit of analysis indirectly.

Units of analysis

Whatever it is that a researcher describes or explains and then draws inferences about The central focus of a study or an analysis. The what or who being studied. In social science research, the most typical units of analysis are individual ppl; but social artifacts like books or movies and groups or populations are often studied as well; Units of analysis- those things we examine in order to create summary descriptions of all such units and to explain differences among them. Some studies try to describe or explain more than one unit of analysis

False Criteria for Nomothetic Causality

When researchers say one variable causes another, they do not necessarily mean to suggest complete causation (In social phenomena, an effect rarely has only one cause. An effect is, typically, the result of a number of causes), to account for exceptional cases (These cases do not disconfirm a causal pattern), or to claim that the causation exists in a majority of cases (If 35% of respondents answered a questionnaire in a particular way, that is still significant even if it is not the majority).

Intervening Relationship

Where the independent variable is causing the effect in the dependent variable through something else in the form of A=B B=C. For example, perhaps results of a study showed that drinking leads to drunk texting. Drinking (A) does not directly cause drunk texting (C); Drinking causes lowered inhibitions (B) which leads to drunk texting. Drinking causes drunk texting through lowered inhibitions.

The relationships between the variables must be non-spurious (third criteria for nomothetic causation)

You should guard against spurious relationships. The incorrect inference of a causal relationship between two variables where the relationship is in reality only accidental coincidence or the presence of a certain third, unseen factor

Ecological fallacy

generalizing from the whole of the group to the individual behavior; Assuming that something learned about an ecological unit (e.g., a group) says something about the individuals that make up that unit; Drawing conclusions about individuals based solely on the observations of groups

Panel Conditioning

the effect of repeatedly measuring variables on members of a panel study