Lecture
The interrelation of transitions between the theoretical hypothesis and empirically loaded hypotheses, as well as the logic of establishing an experimental fact, make it possible to incorporate particular empirical data into a scheme of causal inference. It is necessary not only to formulate hypotheses competently but also to be oriented within their system: in the relationships between:
- theoretical and experimental hypotheses;
- the experimental hypothesis and the counter-hypothesis;
- empirically loaded and statistical hypotheses, etc.
The aim of any experiment is to test the research («working», in Teplov’s terms) hypothesis, which is directed at a plausible explanation of the possibility of seeing in the controlled factor an «influence» that determines changes in the DV.
The experimental hypothesis (EH) is a statement about the type of relationship between the IV and the DV, interpreted on the basis of a theoretical hypothesis or a set of conditions within the experimental model. This is a causal hypothesis in which the IV is regarded as a causally acting factor.
An EH is always a positive statement.
It is opposed by the counter-hypothesis (CH).
The counter-hypothesis is a hypothesis that denies the relationship assumed in the EH or is opposite to it in direction.
The EH and CH are empirically loaded; the constructs representing the DV and IV in the EH are operationalized.
There are two levels at which psychological hypotheses are formulated: the theoretical level (the theoretical hypothesis, TH) and the empirical level (here the consequences of the TH are made concrete in the EH with specified IVs and DVs and an assumed type of relationship between them).
In an experiment with practical aims, the TH may be absent.
In an experiment with scientific aims: the starting point of consideration is not the working hypothesis but the TH; then, by means of it, the EH is made explicit, which automatically generates the CH.
TH→EH→CH;
The EH is placed under such conditions of testing that the chances of obtaining data both «for» and «against» are equal. In order to choose between the EH and the CH, it is necessary to take into account the results of statistical decisions (inferences).
Statistical hypotheses are hypotheses about the sample values of psychological indicators (indicators of the DV).
On their basis a statement is made about accepting or rejecting the statistical null hypothesis, which makes it possible to return to the evaluation of psychological hypotheses.
A distinction is drawn between the formulations of the null hypothesis H0 – as a hypothesis about the absence of differences between the values of a variable under different conditions (absence of a relationship) – and directional hypotheses H1.
H0 may or may not be rejected, but it cannot be proven.
One and the same empirical regularity may admit different causal interpretations, proceeding from different theoretical premises. These other explanations are the «third» (with respect to the EH) competing hypothesis. There may be several such hypotheses.
|
Levels of hypotheses |
Types of hypotheses |
Possible «third» hypotheses |
|
Theoretical |
TH |
A competing or another theoretical one |
|
Empirical |
EH and CH |
Competing, based on identifying «third» variables |
|
Statistical |
H0 and H1 |
|
Transitions between levels, taking into account the old (Fisher) and new (Pearson) traditions.
|
Levels of hypotheses |
Decision-making about the hypothesis |
||
|
Empirical hypotheses |
The EH is accepted |
Neither hypothesis is accepted; search for a third one |
The CH is accepted |
|
Statistical decisions (old tradition) |
Reject H0 at the significance level p=0.05 |
No choice can be made between H0 and H1 |
Do not reject H0 at the significance level p=0.05 |
|
Statistical decisions (new) |
The effect corresponds to H1 (a=0.05) |
No choice can be made between H1 and H2 |
The effect corresponds to H2 (a=0.05) |
Proof by contradiction is a norm that is realized at both levels of transition (from the statistical to the EH, from the EH to the TH), but it involves the specific features of its substantive framing.
Comments