Conceptualization and Operationalization

4) Conceptualization

* Concepts: mental images we use as summary devices for bringing together observations and experiences that seem to have something in common

* Conceptualization: the process of specifying the vague mental imagery of our concept and sorting out the kinds of observations and measurements that will be appropriate for our research.

* Conceptualization means drawing boundaries around terms to make them tangible. Examples: personality, terrorism, violence, democracy, international law…etc.

How to conceptualize: First define and then operationalize (which is one way of defining concept).

1- definition by synonym: the weakest.. Ex. authoritarianism is dicatatorship.

2- definition by classification: Ex. differentiating between totalitarian, traditional authoritarian, competitive authoritarian, emerging democracies and consolidated democracies.

3- definition by enumeration: Ex. listing the names of Arab countries.

4- definition by example, Ex. Consolidated democracy such as Britain…

5- definition by exclusion, Ex. By civil society I do not mean political parties.

These all are important steps and tools to reach:

6- definition by indicators (Operationalization). Ex. "intelligence" is often defined as the score one gets on an intelligence test.

5) Operationalization

· It is the process of developing operational definition; that is a clear protocol (list of steps) for how to quantitatively measure the concept (i.e. to change the concept into a variable with measurable indicators).

· The protocol must be specific and precise enough so that someone else could use your operational definition and obtain the same results you did.

· To sum up, operationalization means to add measurable indicators to concepts to make them measurable variables.

Exercise: define and operationalize religiosity, political participation, and social capital.

How to conceptualize?

Take Social Capital as a case study.

Whereas physical capital refers to physical objects and human capital refers to the properties of individuals, social capital refers to connections among individuals – social networks and the norms of reciprocity and trustworthiness that arise from them. In that sense social capital is closely related to what some have called “civic virtue.” The difference is that “social capital” calls attention to the fact that civic virtue is most powerful when embedded in a sense network of reciprocal social relations. A society of many virtuous but isolated individuals is not necessarily rich in social capital. (Putnam 2000: 19)

Social capital refers to the institutions, relationships, and norms that shape the quality and quantity of a society's social interactions... Social capital is not just the sum of the institutions which underpin a society – it is the glue that holds them together. (The World Bank 1999)

Social capital consists of the stock of active connections among people: the trust, mutual understanding, and shared values and behaviors that bind the members of human networks and communities and make cooperative action possible. (Cohen and Prusak 2001: 4)

6) Sampling and Data Collection

* Primary and secondary data

There is a basic distinction in data collection between primary and secondary data.

Primary Data	Secondary Data
Primary data are data collected by the immediate user(s) of the data expressly for the experiment or survey being conducted. It is this data that we will normally be referring to when we talk about "collecting data".	By contrast, secondary data refers to any data collected by a person or organization other than the user(s) of the data. EX. Visit the CIA Fact book to collect information about country's gdp, literacy rate and the like.

* Uses of secondary data

What are the advantage of using secondary data? In other words, why don't researchers always collect their own data? There are actually several very good reasons why secondary data are used:

· Secondary data may provide a context (geographic, temporal, social) for primary data. This allow us to see where out primary data 'fit in' to the larger scheme of things.

· Secondary data may provide validation for primary data, whereby the Secondary data allow us to assess the quality and consistency of the primary data.

· Secondary data may act as a substitute for primary data. In some situations we may simply not be able to collect data, for reasons of access, cost, or time; or the data have been collected once and to repeat the collection process would be undesirable.

The most famous techniques of data collection in Political Science

Anthropological

Behavioralist

Historical

Field work and detailed observations by participation.

Ex. James Scot, Weapons of the Weak

The best way of understanding the present is by observing how individuals and institutions interpret the past (history).

Survey, personal interviews, focus group discussions, experimentations, content analysis (of news papers.)

The best way of understanding the present is by asking people to tell us how they understand the past (history)

Exploring historical memoirs, census, autobiographies.

The best way of understanding the past/present is by searching for how those who lived in the past perceived their past (history)

7- Data Analysis

1- Determine the variables.

2- Formulate the theory-based hypotheses.

3- Test the hypotheses.

A. Types of Variables

Variables: the specific concepts or theoretical constructs we are investigating. Variables are anything that we can measure … that vary.

1- Independent variable: A variable that is postulated to explain another variables (Ex. personal income)

2- Dependent variable: A variable being explained (Ex. voting for George W. Bush in 2004 elections)

3- Extraneous variables: it represent alternative relationships for relationships that are observed between two variables but

A. no “scientific” claim (based on your literature review) can be made about a causal relationship (Ex. one's foot size or the number of hours playing tennis a week and his/her voting behavior).

B. This indicator cannot be a valid and/or reliable measurement of any other variable of interest. Example: the # of bathrooms does not carry any significance in itself to explain why people vote or do not vote for Bush. But, it may be used to indicate the socio-economic status (SES).

4- Control variables: An independent variable that is held constant in an attempt to further clarify the relationship between two other variables. (Example, to isolate the impact of gender, one examines how rich and poor women vote. In other words, we hold the gender variable constant by ignoring/isolating/controlling the effect of being man).

5- Intervening variable: A variable that is necessary to explain an indirect relationship between an independent and a dependent variable (Ex. one's level of education affects one's level of income that affects one's voting behavior).

B. Formulating Hypotheses

· The statements that postulates the relationship between the independent and dependent variables. This statement should have a theoretical rationale (theory) behind its existence.

l The grammar of a hypothesis

- Independent variables

- The dependent variable

l - Values that variables take on

Hypotheses should be embedded in analytical political theory:

l A theory is a general statement about how the world works that specifies a causal mechanism.

l A hypothesis derived from this theory makes a specific prediction that can be empirically verified. It predicts that one or more causal factors will produce a single effect.

Example:

l My theory is that people will be happy with their political system whenever their basic needs are fulfilled.

l What are some testable hypotheses that spring forth from this theory?

l “Countries where citizens are healthy or wealthy will be less likely to experience successful revolutions.”

What are the independent and dependent variables?

l Independent variables:

l Wealth, measured by 1980 per capita GDP.

l Health, measured by 1980 presence or absence of universal care system.

l Dependent variable:

l Presence or absence of successful revolution since 1980

What values can these variables take? Think!

Have a critical view on the following table.

Country	GDP per capita $	Universal Health System	# Revolts since 1980	# of times winning football World Cup	# of elephants a country has	# of K-Mart Stores in the country
USA	2,5000	0	0	2	1,000	1,500
Nigeria	3,000	0	4	3	80,000	0
Britain	17,000	1	0	1	800	700
Pakistan	4,000	0	3	0	15,000	0
Cuba	7,000	1	1	0	120	0

What variables should be kept?

Types of Relationships between Variables

· Positive: education and income.

· Negative: Age and support for gay marriage

· Curvilinear: Age and income

Examples: Determine the independent and dependent variables and the hypothesized (best hunch) direction.

Republicans are more likely than Democrats to support impeachment.
Highly educated people are less likely than poorly educated people to be unemployed.
As unemployment increases, so does crime.
Age and knowledge go hand in hand.
Among teenagers, sex (male or female) affects self-esteem.
People with high self-esteem earn better grades than people with low self-esteem.
Drug use and grades are related.

C. Visit the Theory

• When you are done with testing your hypotheses, you have to use induction to compare your research empirical generalizations with your theory.

• Confidence in theory increases if the empirical generalization is consistent with the theory.

• Confidence in theory goes down if the empirical generalization is NOT consistent with the theory.

Two Main Techniques of Data Analysis

Qualitative (Ragin’s case-oriented)

Quantitative (Ragin’s variable-oriented)

The usage of logical mental processes of deduction and induction.

A researcher should avoid logical fallacies while performing their case studies, historical analysis, textual/content analysis

The main advantages:

1- It allows the researchers to get as close as possible to their data and become familiar with the cases they study.

2- It pays attention to deviant anomalies as much as it pays attention to present/present cases according to Ragin.

3- It does not run into the problem of sampling error or measurement error that the statistical analysis usually runs into.

4- They try to examine cases as wholes not only by examining certain (isolated) variables.

5- Usually this research is flexible enough not to put a prior assumptions that would limit the variables that we should examine.

6- As a result, it cares more about complexity and validity than on generalizability and developing law-like statements.

Statistical: correlation, regression, Bayesian, logit, factor analysis, and ecological inference.

A researcher should check and correct for the violations of the statistical assumptions such as heteroskedasticity, autocorrelation, and multicollinearity.

The main advantages:

1- It allows for studying more than a small number of cases and thus test on a large scale the validity of claims made at the narrow-scale of small-N studies.

2- Quantification urges precision and making use of available data that usually were used on the descriptive level.

3- It allows for considering different competing explanations; a criticism that is usually mounted at case-study.

4- It allows researchers to study cases that they are not specialized in by resorting to data-banks.

5- It allows for testing previous results by drawing other samples, controlling for variables and adding more data.

6- As a result, it care more about general results and law-like statements even at the expense of complexity and validity.

They would criticize variable-oriented statistical analysis as it overlooks that fact that the relationship between the independent variables is not additative but rather it is an interaction which makes each case unique.

They respond that, there are now more techniques that can take into consideration the qualitative data such as interaction models, splitting the sample and creating sub-population and Bayes analysis.

Conclusion: Both quantitative and qualitative methods should always be interwoven together as much as we can. A good example is given by Wiarda when he attempts to show that aggregate statistical (quantitative) analysis may mislead us if it does not entail enough qualitative examination of the case studies. If we use aggregate statistical data to study the correlation between economic growth and democratization for a long period of time, we may reach some misleading conclusions. For example Nicaragua under the dictator Anastasio Somoza was sometimes listed as “over-democratized” for its level of economic development because one of the indices used to measure democratization – the presence of opposition members in the congress – was consistently high in that country. What aggregate quantitative analysis cannot reveal is that in Somoza’s non-democratic regime the constitution required one-third of the legislature to come from opposition parties so that the “dictator could portray his regime as more democratic than it really was.” (Wiarda, 10) This example gives credit to Tarrow’s argument of triangulation of different methods on the problem (P. 473).