The questionnaire (Table 2) was constructed especially for this case and proved to be both reliable (internal consistency) and valid (concurrent, convergent, discriminant, and statistical conclusion validity; data not shown in detail; see also statistical analysis section) for the Greek population used.
In conclusion, validity results based on scores measured in low-stakes contexts cannot simply be generalized to high-stakes contexts.
This design is distinguished from quasiexperiments in that the plausible threats to internal validity, regression and maturation, as well as the threats to statistical conclusion validity (Cook and Campbell 1979) must be controlled by a statistical model.
Donald Campbell and Julian Stanley invented the term quasi-experiment, described many variants of these designs, invented a validity typology (statistical conclusion validity, internal validity, construct validity, external validity) and threats to validity to help evaluate causal inferences from these designs, and emphasized the use of design features to rule out or diagnose these threats.
Shadish et al. (2002) have listed nine threats to statistical conclusion validity, that are examined in this article, and the strategies for overcoming these threats are identified for researchers designing effective research investigations.
Regarding statistical conclusion validity of quantitative articles, articles having endogeneity threats received significantly fewer citations than did those using a more robust design or an estimation procedure that ensured correct causal estimation.
The classic definition of statistical conclusion validity that is an essential design component for research investigations is given by Cook and Campbell (1979: 37) as the truthfulness of "conclusions about covariation" between treatment and outcome – that is, the degree of accuracy with which researchers can conclude that two variables are related or that a treatment is effective.
Hence, our study has a high conclusion validity.
However, there is a possible threat to statistical conclusion validity.
