Strategy for testing a proposition that expresses a sufficient condition

than the presence of A, it is very unlikely that B is present, or for other reasons than the absence of B it is very unlikely that A is absent. The survey could be used to test a sufficient condition as well. Remember that we define a survey as a study in which a a single population in the real life context is selected, and b scores obtained from this population are analysed in a quantitative manner. For a test in a survey, a population can be selected in which the dependent con- cept B is known to be present or the independent sufficient condition A is known to be absent. If the values of the concepts are unknown, any population could be selected from the domain. It is tested in this popu- lation whether the frequency of occurrences of instances with the val- ues A presentB absent is zero as expected if the proposition is true in the population or is very small according to a “pragmatic determin- ist” criterion, as discussed in Box 8. The hypothesis is rejected if the proportion of instances with the values A presentB absent is larger than zero or larger than the proportion specified. Such a survey might seem an efficient way of testing a sufficient condition, because it is an efficient way of computing the proportion of the instances in which the proposition is not correct. We have classi- fied the survey as the third-best strategy for testing a sufficient condi- tion for the following two reasons. 1. When the survey strategy as discussed here is used for testing a sufficient condition it is one test in the set of all instances of a sample from the selected population. This strategy for testing the sufficient condition is comparable with a case study with many parallel replications at the same time, in which for each instance it is determined whether it is or is not an instance with the values A presentB absent. In sec- tion 3.2.2 “Replication” we showed that parallel replication may not be efficient. If a rejection of the hypothesis is found in a number of instances, this might be a reason to stop fur- ther testing of the proposition. But, in the survey strategy, scores of all instances of the population or of the sample must be known because the test is by definition conducted in the entire sample of the population. The parallel single case study, thus, is much more cost effective in terms of measure- ment costs. 2. The survey tests the proposition in only one population, which is selected from all possible populations in the domain. Other surveys are needed to replicate the test in other parts of the domain, which again implies measurement costs. If the same number of instances would be observed in a serial single case study, these could be selected much more purposively from all parts of the domain. The serial single case study, thus, is considerably more flexible and efficient.

4.4.2 Strategy for testing a proposition that expresses a necessary condition

A proposition that expresses a deterministic necessary condition implies that for each single instance in the domain the proposition is true according to the theory. Again, this means that the proposition can be tested in a single instance. A proposition with a necessary condition can be confirmed with an experiment in a situation where A and B are both present and by taking away the condition A and observing whether the effect B disappears. If conducting an experiment is not feasible, the best strategy for test- ing a necessary condition is the single case study. One instance of the object of study a case in which effect B is present is selected, and it is observed whether condition A is present or not. If not, then the hypothesis is rejected. Referring to Figure 4.2, the hypothesis is rejected if the case is located in the upper left cell, because according to the hypothesis that cell must stay empty. An alternative test is that one instance of the object of study a case in which condition A is absent is selected, and it is observed whether effect B is present or not. If B is present, then the hypothesis is rejected. Referring to Figure 4.2, the hypothesis is rejected if the case is located in the upper left cell because according to the hypothesis that cell must stay empty. Again, as with testing for a sufficient condition, it is not possible to confirm the correctness of the proposition for all instances of the domain without repeating the test in all of them, but finding one instance in which the proposition is rejected is sufficient for conclud- ing that the proposition is not correct for at least one instance from the domain to which it was assumed to apply. As with testing for a suf- ficient condition, a failure to find rejections of the hypotheses in many different attempts, particularly in “least likely” cases i.e. in instances in which B could be expected to occur anyway, even without A provides some confidence that the proposition might be correct for the domain in which it was tested. The survey might be used to test a necessary condition as well. For a test in a survey, a population can be selected in which the necessary condition A is known to be absent or the dependent concept B is known to be present. If the values of the concepts are unknown, any population could be selected from the domain. It is tested in this popu- lation whether the frequency of occurrences of instances with the val- ues A absentB present is zero as expected if the proposition is true in the population or is very small according to a “pragmatic determin- ist” criterion, as discussed in Box 8. The hypothesis is rejected if the proportion of instances with the values A absentB present is larger than zero or larger than the proportion specified. The same argument about inefficiency of the survey as discussed above for the use of the survey for testing a sufficient condition applies here as well.

4.4.3 Strategy for testing a proposition that expresses a deterministic relation

A proposition that expresses a deterministic relation implies that for each single instance in the domain the proposition is true according to the theory. This means that the proposition can be tested in a single instance. The preferred strategy for testing a deterministic relation is the experiment. In such an experiment it must be demonstrated that each change in the value of the independent concept results in a predicted change in the value of the dependent concept. Depending on whether condition A can be administered in different dosages, the experiment could either be cross-sectional in which different values of A are administered to different groups or longitudinal in which the value of A is, for instance, gradually increased over time. The hypothesis is confirmed if the effect B increases according to the prediction. If an experiment is not feasible, the longitudinal single case study or the comparative case study is the second-best strategy. In the longitu- dinal single case study one instance is selected for measurement of both the independent and the dependent concept over time. It is assessed for each measurement point separately whether the value of dependent concept corresponds to the expected value. In the com- parative case study, two or more instances are selected each with a different value of the independent concept and the value of the dependent concept is observed, or one instance is selected for meas- urement of both the independent and the dependent concept over time. It is assessed for each measurement point separately whether the value of the dependent variable corresponds to the expected value.