qualitative comparative analysis

Post on 23-Feb-2016

175 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

DESCRIPTION

Qualitative Comparative Analysis . What, When and How? Dumitrela Negur ă BA. Qualitative Comparative Analysis (QCA). Introduced by Charles Ragin in 1987, when stumbling upon the causal inference problems generated by a small sample - PowerPoint PPT Presentation

TRANSCRIPT

Qualitative Comparative Analysis

What, When and How?

Dumitrela Negură BA

Introduced by Charles Ragin in 1987, when stumbling upon the causal inference problems generated by a small sample

Represents a method that bridges qualitative and quantitative analysis

Why? Because it is difficult to do in-depth qualitative work with sets larger than 15 (although not impossible) and is not very meaningful to do traditional statistical approaches on sets this small

Qualitative Comparative Analysis (QCA)

Most aspects of QCA require familiarity with cases and in-depth knowledge of the theory

With QCA, it is possible to assess causation that is very complex, involving different combinations of causal conditions capable of generating the same outcome

It is used in comparative case-oriented and in small scale research, for studying a small-to-moderate number of cases in which a specific outcome has occurred, compared with those where it has not

It is very useful when you have small samples (N=8 to N=200 or N=5 to N=50)

Used in : sociology, psychology, political science and history but can be applied to health related research

When do we use it ?

QCA uses as units of analysis crisp and fuzzy sets and subsets

How?

QCA was developed originally for the analysis of configurations of crisp set memberships (conventional Boolean sets)

With crisp sets, each case is assigned one of two possible membership scores in each set included in a study: 1 (yes/ presence) or 0 (no/ absence)

Crisp sets

Fuzzy sets( fs/QCA) solve the problem of trying to force-fit cases into one of two categories

Fuzzy sets can have three or more categories (any value between 0 and 1):

1.00 = fully in 0.80 = mostly in 0.60 = more in than out0.40 = more out than in0.20 = mostly out0.00 = fully out

! Are not well suited for conventional truth table analysis !

Fuzzy sets

Crisp vs. Fuzzy sets

The simple way is to construct truth tables ( used only for crisp sets) and use Boolean algebra, considering all the logical combination of the causal conditions

The three basic Boolean operators are:o logical OR (+)o logical AND (*)o logical NOT (replacing the upper case letter with a lower

case letter) A dash symbol [-] represents the “don’t care” value for a

given binary variable, meaning it can be either present (1) or absent (0)

The arrow [→] is used to express the link between a set of conditions

For example: A+B *C-> Y or a+B*c->y ( where Y is the outcome)

Crisp-set analysis

Truth tables list the logically possible combinations of causal conditions and the outcome associated with each combination

Truth tables help us to see clearly the similarities, differences and contradictions between cases

The number of combination is a geometric function of the number of causal conditions (number of causal combinations = , where k is the number of causal conditions)

Truth tables

Causal relations are interpreted in terms of necessary and sufficient conditions

With necessity, the outcome is a subset of the causal condition

With sufficiency, the causal condition is a subset of the outcome

Boolean logic is used to reduce the table to a few statements indicating necessary and sufficient conditions and their combinations

 

Cases

Genes and family history

Unhealthy food

Inactive lifestyle Environment Health

conditionsOutcome :

Obesity

1 1 0 0 0 0 12 0 1 1 0 1 13 1 1 1 0 1 14 0 0 0 1 0 05 0 1 0 0 0 06 0 0 0 1 0 07 0 1 1 1 1 18 0 1 0 1 0 09 1 0 0 0 0 1

10 1 1 1 0 1 1

Example:

The number of combinations for this example will be

 

Cases

Genes and family

history (G)Unhealthy food (U)

Inactive lifestyle (L) Environment (E) Health

conditions (H)Outcome : Obesity (O)

1,9 1 0 0 0 0 12 0 1 1 0 1 13, 10 1 1 1 0 1 14,6 0 0 0 1 0 05 0 0 0 0 0 07 0 1 1 1 1 18 0 1 0 1 0 0

Truth table: configuration and minimization

This means that we have these possible combinations: G*u*l*e*h + g*U*L*e*H + G*U*L*e*H + g*U*L*E*H -> O

g*u*l*E*h + g*u*l*e*h +g*U*l*E*h -> o

For example G is a sufficient condition and U is necessary but not sufficient for the outcome(O).

Because the truth tables can be very complex because of their size, a specialized software can be used

The software can generate the truth table and also analyzes fuzzy sets

Software

For crisp-set analysis: fs/QCA TOSMANA QCA 3.0

For fuzzy-set analysis: fs/QCA

Free and user friendly softwares

Regression analysis vs. QCA

QCA offers an alternative approach, bridging the qualitative and quantitative methods and it’s used for small scale research

Used for assessing causation

Uses theory-set relationships

Not hard to use but it demands good knowledge of theory and cases

To summarize:

Thank you

top related