Types of hypothesis in management research

Theoretically there should be only one type of hypothesis, that is the research hypothesis — the basis of your investigation. However, because of the conventions in scientific enquiries and because of the wording used in the construction of a hypothesis, hypotheses can be classified into several types. Broadly, there are two categories of hypothesis:

  1. research hypotheses;
  2. alternate hypotheses.

The formulation of an alternate hypothesis is a convention in scientific circles. Its main function is to explicitly specify the relationship that will be considered as true in case the research hypothesis proves to be wrong. In a way, an alternate hypothesis is the opposite of the research hypothesis. Conventionally, a null hypothesis, or hypothesis of no difference, is formulated as an alternate hypothesis.

Let us take an example. Suppose you want to test the effect that different combinations of maternal and child health services (MCH) and nutritional supplements (NS) have on the infant mortality rate. To test this, a two-by-two factorial experimental design is adopted (see Figure 6.2). There are several ways of formulating a hypothesis. For example:

  1. There will be no difference in the level of infant mortality among the different treatment modalities.
  2. The MOH and Ns treatment groups will register a greater decline in infant mortality than the only MOH treatment group, the only NS treatment group or the control group.
  3. Infant mortality in the MCH treatment group will reach a level of 30/1000 over five years.
  4. Decline in the infant mortality rate will be three times greater in the MCH treatment group than in the NS group only over five years.

Let us take another example. Suppose you want to study the smoking pattern in a commu­nity in relation to gender differentials. The following hypotheses could be constructed:

  1. There is no significant difference in the proportion of male and female smokers in the study population.
  2. A greater proportion of females than males are smokers in the study population.
  3. a total of 60 per cent of females and 30 per cent of males in the study population are smokers.
  4. There are twice as many female smokers as male smokers in the study population.

In both sets of examples, the way the first hypothesis has been formulated indicates that there is no difference either in the extent of the impact of different treatment modalities on the infant mortality rate or in the proportion of male and female smokers. When you construct a hypothesis stipulating that there is no difference between two situations, groups, outcomes, or the prevalence of a condition or phenomenon, this is called a null hypothesis and is usually written as H0.

The second hypothesis in each example implies that there is a difference either in the extent of the impact of different treatment modalities on infant mortality or in the proportion of male and female smokers among the population, though the extent of the difference is not specified. A hypothesis in which a researcher stipulates that there will be a difference but does not specify its magnitude is called a hypothesis of difference.

A researcher may have enough knowledge about the smoking behaviour of the community or the treatment programme and its likely outcomes to speculate almost the exact prevalence of the situation or the outcome of a treatment programme in quantitative units. Examine the third hypothesis in both sets of examples: the level of infant mortality is 30/1000 and the pro­portion of female and male smokers is 60 and 30 per cent respectively. This type of hypothesis is known as a hypothesis of point-prevalence.

The fourth hypothesis in both sets of examples speculates a relationship between the impact of different combinations of MCH and NS programmes on the dependent variable (infant mortality) or the relationship between the prevalence of a phenomenon (smoking) among different populations (male and female). This type of hypothesis stipulates the extent of the relationship in terms of the effect of different treatment groups on the dependent variable (‘three times greater in the MCH treatment group than in the NS group only over five years’) or the prevalence of a phenomenon in different population groups (‘twice as many female as male smokers’). This type of hypothesis is called a hypothesis of association.

Note that in Figure 6.3 the null hypothesis is also classified as a hypothesis of no differ­ence under ‘Research hypothesis’. Any type of hypothesis, including a null hypothesis, can become the basis of an enquiry. When a null hypothesis becomes the basis of an investigation, it becomes a research hypothesis.

Source: Kumar Ranjit (2012), Research methodology: a step-by-step guide for beginners, SAGE Publications Ltd; Third edition.

One thought on “Types of hypothesis in management research

Leave a Reply

Your email address will not be published. Required fields are marked *