1. Scientific inference starts with a theoretical statement, an element of a theory, which says one class of phenomena will be connected in a certain way with another phenomena.

2. EXAMPLE Protestant versus Catholic
Durkheim’s theory of egoistic suicide: “A higher degree of individualism in a social group causes a higher rate of suicide in that group.”
Where “individualism” is a variable that means: “the degree to which all activities of the person are controlled by well-defined norms enforced regularly and effectively by people in the environment, as opposed to morality determined by the individual himself by his own decisions.”
Single male versus married male with children
Protestant versus Catholic
Jews versus other social groups in France

3. Concepts “Individualism” a property of groups inferred indirectly
“Suicide rate” lower level of abstraction and is directly observable

4. Based on logical deduction and operational definition of concepts we derive an empirical statement.
An empirical statement is one that is implied logically from the theoretical statements.
“If we make such and such an observation, they will have such and such results.”

5. Some examples of Durkheim’s empirical statements:
Protestants in France will have a higher suicide rate than Catholics in France.
Protestant regions of German provinces will have higher rates of suicide than Catholic regions.
Married men in France will have lower rates of suicide than single men and will have even lower rates if they have children.
Men who practice the free professions and generally well-educated men will have higher rates of suicide than workers or less educated people.
In times of parliamentary crisis, the suicide rates will go down in France
Stating the meaning of the concepts in terms of observations.

6. Testing Theories with Observations
Situation I Situation II A B A B
B false B true A false A more credible
Where A is the theory and B is one of its empirical consequences. Arrow refers to “implies”.
For example, if Durkheim’s theory implied Protestants ought
to have a higher suicide rate in France, and they in fact have a
lower rate, then his theory is false.

7. Multiple Tests of Theories
Situation III
A B1, B2, B3
B1, B2, B3 all true
A substantially more credible
Example where B1 is Protestant countries will have higher suicide rates than Catholic countries, B2 is Protestant regions will have higher suicide rates than Catholic regions, and B3 is Protestants in France will have higher suicide rates than Catholics.
A multiple test of a theory is more convincing than a single test!

8. A B1, B2, B3 B1, B2, B3 all true and different A much more credible Example where B1, B2, B3 are different from the regional religious difference above in that it is shown that men with children have a lower suicide rate than bachelors and men without children If the theory can imply such different kinds of empirical statements as one about marital status, then we feel intuitively that the theory has been subjected to a tougher test than it has if we merely repeat observations similar to those we have already made.

9. In general , even before any research begins there are more than likely a number of alternative explanations, or alternative theories, that one might hold about what goes on in the world. Some of these alternative theories may have been thought of before while others have yet to be formulated.

10. Situation II, Reformulated (Situation before testing B1)
For any given observation which is an implication of A, say B1, there will be some of the possible theories that will imply not-B1. If we demonstrate B1, these alternative theories are falsified. This yields fewer alternative possible theories.
Situation II, Reformulated (Situation before testing B1)
A or (C, D ,E,...,Q, R , S,...)
A B1
(C, E,...,Q, S...) B1
D, R NOT-B1
B1 true
D,R false (by classical logic)
A OR (C, E,...,Q, S,...)
A more credible (but also C, E,...,Q, S,...more credible)
That is, with the test B1 we eliminate D and R from among the possible alternatives, making all of those which imply B1 more credible. Among those theories made more credible by this elimination of competitors is our own theory, A.
Situations III and IV can be reformulated in this same way.

11. Thus the basic logical process of science is the elimination of alternative theories (both those we know and those we do not know) by investigating as many of the empirical consequences of each theory as is practical, always trying for the greatest possible variety in the implications tested.

12. EXAMPLE: The Ideology of Folk Resource Management
Theoretical Statement: Contrary to the Tragedy of the commons notion, “traditional” individuals can restrain their exploitation of harvestable resources in order to “conserve the resource” in that they display a conservation ethic. Even in the presence of overexploitation, “traditional” or “indigenous” fishers will restrain their harvesting behavior in order to conserve the resource.

13. Empirical Statements
The fishers of the rural village of Santo Domingo (“traditional” fishers) will use less destructive fishing gear than the neighboring nontraditional fishers form the urban area of Oro Grande.
The fishers of the rural village of Santo Domingo (“traditional” fishers) will be more selective about the size of fish they harvest, taking only larger fish in order to allow the smaller fish to grow, while the nontraditional fishers of Oro Grande will not be selective about the size of fish harvested.

14. The observations:
Fishers of Santo Domingo used spear guns and not nets while the fishers of Oro Grande used nets.
Fishers of Santo Domingo selected only larger fish on the outer reefs while fishers of Oro Grande fished both inner and outer reefs and harvested many small fish.

15. Conclusions
Traditional fishers displayed a conservation ethic in that they employed less destructive gear that is more discriminating in terms of size and species. They stated that they were selecting only the larger fish so that the smaller fish could grow and that they were not fishing the inner reef in order to allow it to recover from over fishing.
Traditional fishers target adult fish more frequently than juveniles, thus allowing them to grow up, while nontraditional fishers harvest indiscrimenantly leading to “ruin for all”.

16. Competing Alternative
Selection of larger fish by traditional fishers is due to market demand for larger fish and fishing of outer reef since over fishing of the inner reef has caused a lack of larger target species in that region Other problems
Possible erroneous biological assumptions
Lack of proof that nontraditional fisher practices are ecologically and biologically damaging
Ideological blinders
Samples

17. The Point of All This There exists two alternative theories that we always have to regard as possible explanation of any given set of observations:
observations produced by the way the study was designed e.g., sample of observations chosen
observations produced by a large set of small influences operating in different directions e.g., large number of small measurement errors We always want to reject evidence if it can be explained either by the design of the research or by a small number of small, unorganized causes.