1. Introduction
A hypothesis is a claim, an assumption, a guess, a prediction, or an allegation. Every day, we form hypotheses. The majority of hypotheses we form daily are educated guesses aimed at explaining why something is happening. For example, if all of the electrical devices in your room suddenly go out, you may come up with a series of hypotheses:
a) There is a widespread power outage.
b) Your prepaid credit on your meter has run out.
c) The meter is broken.
All of these hypotheses can be tested. In the first instance, you might go out to ask your neighbours if they have electricity. In the second scenario, you will step outside to check on your prepaid meter. In any event, any hypothesis must be tested before it is used or accepted. Untested hypotheses are only claims or allegations and cannot be utilized to make decisions.
A hypothesis is typically defined in research circles as anticipation of the outcome of a study. It is an assertion concerning the expected link between variables in a study, often the independent and dependent variables. In a study on the impact of customer service on customer satisfaction, for example, we may provide our prediction of the outcome via a hypothesis:
Customer service will have a positive impact on customer satisfaction at the end of the study.
This hypothesis has:
a) Linked the study's variables (customer service, customer satisfaction)
b) Given the study's anticipated outcome
Their expectations will now direct the research.
It is important to note that once a hypothesis is made, it must be tested. Testing a hypothesis is gathering data to confirm that the assertion (hypothesis) is correct. When a hypothesis is discovered to be true, it is accepted; when discovered to be erroneous, it is rejected.
2. Types of Hypotheses
There are different types of hypotheses. Because it is entirely research-based, this article will explore hypothesis categories for their utility in research. In this vein, hypotheses will be classified according to whether they are stated as:
- Research hypothesis
- Statistical hypothesis
2.1 Research Hypothesis
A research hypothesis, also known as a declarative, scientific, or substantive hypothesis, is a statement that describes the expected relationships between or among two or more variables. This relationship is what the researcher hopes to confirm through data collection.
Examples
- The amount of rainfall will have a positive influence on maize yield.
- Students who take extra classes perform better on the end-of-term exam than students who do not.
- Adolescents who were sexually abused as children have a higher risk of depression and suicide than adolescents who have never been sexually abused.
Furthermore, research hypotheses might be directional (indicating the exact direction of the change) or non-directional.
2.1.1 Directional Research Hypothesis
A directional research hypothesis describes the expected direction of change at the end of the study. The direction can be upwards or downwards, negative or positive, and more or less. In this case, the researcher anticipates not only the existence of a relationship but also its nature.
This prediction is often based on a previous study, accepted theory, substantial experience, or relevant literature.
This form of research hypothesis is employed when the researcher expects a specific study outcome. Here are several examples:
- Students from private schools outperform students from public schools in the Basic Certificate Examinations.
- Customer satisfaction will improve as a result of improved customer service.
- The experimental group's systolic blood pressure will be lower than the control group's.
- There is a positive relationship between the amount of time spent on tasks/activities and how well the task is completed.
2.1.2 Non-Directional Research Hypothesis
A non-directional hypothesis, on the other hand, does not specify or express the direction of the relationship. A hypothesis of this type predicts that two or more variables are related but makes no predictions about the nature of the relationship. The researcher feels there will be a difference in the connection, but she is unsure what it will be.
Non-directional hypotheses include the following:
- There will be a difference by gender in an IQ test in a sixth-grade class.
- Customer service will have an impact on customer satisfaction
- There will be a difference in the expression of satisfaction with the counselling process between students who get centred treatment and students who receive standard therapy.
2.2 Statistical Hypothesis
A statistical hypothesis is a statement that describes a relationship or difference that can be statistically tested. The research hypothesis is translated into these hypotheses. It enables the researcher to rewrite the study hypothesis in such a way that statistical procedures can be applied to it.
A statistical hypothesis is simply a claim made about a population that can be tested by choosing a random sample from the population of interest. A statistical hypothesis is simply a claim made about a population that can be tested by choosing a random sample from the population of interest.
There are two forms of statistical hypotheses:
- The null hypothesis (Ho, pronounced “H-nought”)
- The alternate hypothesis (H1 or Ha)
2.2.1 Null Hypothesis
A null hypothesis states that there is no relationship (or difference) between variables and that any relationship found will be a chance relationship, not a true one.
Null means nothing, or zero.
The null hypothesis is tested and its acceptance or non-acceptance provides support or no support for the research hypothesis. Examples of null hypotheses are:
- Interest rates will have no effect on inflation
- There will be no difference by gender in an IQ test in a sixth-grade class.
NB: In a law court, the judge begins with the null hypothesis: He is NOT guilty. The prosecution would now have to prove, with facts (data), that the null hypothesis is wrong (he is guilty).
2.2.2 Alternate Hypothesis
The alternate hypothesis is identical to the research hypothesis given in section 2.1. This means it can be a directed or non-directional hypothesis. Here are several examples:
- Counselees who receive client-centred treatment are happier with the counselling process than those who receive traditional therapy.
- Customer service will have an impact on customer satisfaction
It is worth noting that the null and alternate hypotheses are presented simultaneously. E.g.,
Null hypothesis: temperature has no effect on the eating habits of rats
Alternate hypothesis: temperature will have an effect on the eating habits of rats
Mathematically, they may be stated as:
Ho: µ ≠ 10 (assuming the population mean, µ is 10)
H1: µ = 10
3. Important Qualities of a Good Hypothesis
a) A good hypothesis should usually be testable within a reasonable time frame.
b) The research hypothesis should be testable empirically.
c) The research hypothesis should be derived from the research problem. In general, a problem may have several variables. In this situation, many research hypotheses might be given to highlight the entire expected relationship. For example:
Ho: Bank size, Credit Risk (CR), Operational Efficiency (OE), Asset Management (AM), and Debt Ratio (DR) have no impact on Return on Assets (ROA).
Ho: Bank size, credit risk, operational efficiency, asset management and debt ratio have no impact on Tobin’s Q.
Ho: Independent variables have no impact on Return on Equity (ROE).
d) There are no restrictions on the number of hypotheses that can be stated and investigated.
4. CONCLUSION
You have the option of using either the research hypothesis or the statistical hypothesis. However, for the convenience of testing with statistics, it is sometimes more convenient to employ the statistical hypothesis in quantitative studies.
ABOUT THE AUTHOR
Michael Dabi is a professionally trained Environmental Scientist and Biological Oceanographer. Through writing, he enjoys clarifying complex research and statistical concepts for student comprehension.
Comments