Research questions and hypotheses are two internationally standard and essential components of academic research. This paper will attempt to make a distinction between the two concepts, in the process highlighting important elements within each concept. This paper will begin with an overview of the research process, explain the concepts of hypotheses and research questions and show how these two concepts are related to each other.
In the second part of this paper, emphasis will be brought to bear on hypotheses and their uses in research, providing the reader some examples to illustrate. It will show how hypotheses are linked with other parts of a research study and conclude with the limitations of hypotheses.
The third part of this paper examines the use of research questions in research. It offers examples of research questions, shows how they can be linked with other parts of the study and it then examines their limitations.
The final part of this paper compares and contrasts hypotheses and research questions and it is from this point that the final conclusion is drawn.
A General Overview of Research
Research is defined as “the art of scientific investigation” (Kothari, 2004:p.1). It is the quest to obtain new knowledge or to extend the boundaries of that which is already known. Research, therefore, seeks to discover, explain and predict phenomena through the skilled and systematic manipulation of variables.
The research process starts when the researcher encounters a question or related questions which he or she thinks could and should be answered. When a researchable question comes up in the problem area of the researcher, he or she gathers tools to find out whether the proffered solutions offered by the hypothesis/hypotheses are workable and if an intervention were to be carried out, the likely chances of its success.
In academic research, the researcher is expected to link the research questions with the purpose of their study and their objectives.
Explaining the Concepts
The following are brief explanations of the key terms used in this paper which include hypotheses, research questions and the relationship between them.
A tentative explanation for an observation, phenomenon or scientific problem that can be tested by further investigation
Formally stated questions intended to provide indications about something of interest or relationships between variables in the research process
A quantity or attribute of a construct which is likely to vary
What are Hypotheses?
‘Hypotheses’ is the pluralized form of ‘hypothesis’. It originates from the Greek word hypothesis, which means groundwork, supposition or foundation. A hypothesis therefore from its early origins was used to mean a supposition, or unproven explanation for any given phenomenon.
To the layman, hypotheses are simply conjectures which are proposed ideas to explain facts or observations, or simply stated, an educated guess. An example of this could be “Dar es salaam is hotter than other cities because it has fewer trees.”
Hypotheses are also used to mean expressing an opinion based on incomplete evidence such as when a person says “thinking good thoughts can heal your illnesses, and thinking bad thoughts will make you ill.”
In research, however, a research hypothesis represents a predictive statement, testable by the scientific method of inquiry, which relates an independent variable/s to some dependent variable(s) (Kothari, 2004, p.201). A hypothesis is also defined as ‘a testable proposition about the relationship between two or more concepts’ (Gray 2004, p.7). Another important fact about a hypothesis is that it retains the character of an educated guess until facts are found to confirm or discredit it (Mauch & Park, 2003, p. 136).
When hypothesis are used in research, the ‘proposition’ must be testable. The following are examples of testable hypotheses:
Students who study for one hour a day or more perform better than students who study less than one hour a day
Students who learn using the ‘problem based learning’ methodology retain knowledge for longer than students who learn using traditional methods of learning.
What are Research Questions?
A research question is a formally stated question intended to provide indications about something of interest in the research process. It is a statement which identifies the phenomenon to be studied. A research question, unlike a hypothesis, is not limited to investigating relationships between variables. The research question is a statement framed as a question which may ask about the relationship between variables (such as dependent and independent variables) in a research study, or which may make an interrogative general statement about the area under study.
The Relationship between Hypotheses and Research Questions
While a hypothesis is a formal statement about the relationship between variables, the research question is a formally stated question providing indications about concepts in the study and not limited to investigating relationships between variables.
Hypotheses are often guided by research questions. While some research questions may be tested by hypotheses, others do not require hypotheses testing (see page 10 below). A well thought out and focused research question is expected to lead to a hypothesis showing relationships between two variables.
Hypotheses, although they are important, are not essential for an academic study. There are studies in qualitative research which are focused on theory building which start with a mere research question and end in a series of generalizations which may be made to form a theory. A research question such as the following could fit into this category: “How do members of the Ankole community experience marriage?”
The Use of Hypotheses in Research
The researcher often hypothesizes in order to help give direction to their work. When the researcher does this, he/she makes inferences based on what is known (facts) and observed conditions in reaching a decision on how to approach a study. The researcher may come to a hypothesis from a thorough analysis of the theoretical and factual background of a research problem.
For example, a researcher conducting a study on the relationship between obesity and cancer may discover a trend from reading medical reports which links the consumption of certain types of foods to the onset of cancers. From this angle, the researcher may develop a hypothesis that the consumption of the foods is related to the onset of cancer and worth studying as a specific hypothesis.
Types of hypotheses
A hypothesis may be classified in terms of how it is derived as either inductive or deductive. While inductive reasoning seeks to create general principles from starting with many specific observations or instances, deductive reasoning seeks to create a specific conclusion based on generalizations.
An inductive hypothesis is a hypothesis formed through inductive reasoning from many specific observations to tentative explanations of the causes. An example of inductive reasoning is from observing from Christian history that leaders in the church have been violent, intolerant and prejudiced to making an inductive hypothesis that Christians are all violent, intolerant and prejudiced.
A deductive hypothesis is one which is formed through deductive reasoning which starts with a premise such as a theory, and then forms a conclusion based on that premise. An example of a deductive hypothesis is using the premise “all Nigerians are scammers” – and Niyi is a Nigerian, therefore the conclusion “Niyi is a scammer”.
A hypothesis may also be classified on the basis of its formulation as a research hypothesis which may be directional or non-directional, the null hypothesis, and the statistical hypothesis.
As mentioned earlier research hypotheses are conjectural statements of expected results. They could either be directional or non-directional.
A directional research hypothesis anticipates a specific outcome and states the expected direction of the relation or difference between variables. An example is “a student class using problem-based learning will demonstrate higher achievement than a student class using teacher centered instruction methods.”
A non-directional research hypothesis is one on which an outcome is anticipated but the specific nature of the outcome is unsure, or put simply, it only states that a relationship exists. An example of a non-directional hypothesis is “there will be differences in achievement between a student class using problem-based learning and one using teacher centered instruction methods.”
The null hypothesis is used in statistics and it is simply a hypothesis that no difference or relationship exists between the variables under study. This does not represent the outcome expected by the researcher and is only used for statistical reasons. The null hypothesis is the opposite of the research (or maintained or alternative) hypothesis.
Linking hypotheses to other parts of the study
In the quantitative approach to research, hypotheses are derived from research questions and are used to link the variables of the study. Each hypothesis may be linked to a specific research question. In empirical studies, the testability (the measure to which the hypothesis may be tested) and acceptability of a hypothesis help negate or accept a theory – the theory upon which the assumption is being made. For example, in a study in which the research hypothesis states that “there is a significant positive relationship between lateness to work and employee productivity”, if the researcher were to discover a significant positive relationship between the two variables listed through the study, it would imply that the alternative hypothesis be accepted. Therefore the findings of the research would strengthen any underlying theory linking the concepts of lateness and employee productivity.
Limitations of hypotheses
A hypothesis is restricted in use in empirical research to concepts which are clear and unambiguous. A hypothesis therefore helps the researcher in the formulation of clear research concepts. A hypothesis may not be used for instance to measure concepts which have different meanings for different people. That is why, for example, it is often an elusive endeavor trying to measure concepts such as intelligence which do not lend themselves to clear empirical testing.
Concepts in hypothesis must have indicators with which they can be measured by. The means of measuring indicators or instruments of measurement must be valid and reliable if the results from hypothesis testing are to be relied upon.
A hypothesis is also limited to testing researchable concepts. Concepts such as “the will of God” may not be researchable due to the difficulty of finding valid measures.
Use of Research Questions in Research
Research questions are used in different ways depending on the research approach. For the purpose of this paper, we shall see how they are used in quantitative and qualitative research.
Research questions could be used in quantitative studies to highlight the key variables in the research problem which need to be compared with one another in order to draw a conclusion. When they are so used they may generate a hypothesis for each question as outlined in the cases outlined below. Research questions may also be used in descriptive studies which aim to describe the characteristics of an event, region or community. An example of a research question as used in a descriptive study could be “What candidate are Nigerian voters likely to support for President.”
Research questions, when used in quantitative research, must relate to the problem statement and outline the specific query which the researcher seeks to answer. After the research question(s) is/are stated, the resulting hypothesis is outlined. For example, if the problem of the study is related to why women have higher scores than men on an IQ test in a certain district, one of the research questions may be:
“What percentage of women has higher IQ scores than men in this district?”
In qualitative research such as in “Grounded Theory”, or “Critical Theory” research, research questions usually focus on one concept or idea rather than making assumptions among groups. In this type of research, research questions also are often based on hunches or personal experience. An example of a research question in qualitative research is “How do Kenyan women view domestic violence?”
In qualitative research, research questions may generate data which may generate a pattern for stating a generalized statement such as a hypothesis or theory.
Types of Research Questions
Gray (2004, p. 70) distinguishes between four types of research questions: descriptive, normative, correlative and impact research questions.
Descriptive research questions are about what is happening and may show which methods are being used. An example of a descriptive research question is “what proportion of private universities has a health and safety policy?” Descriptive questions may be answered without the need for a hypothesis as their aim is simply to establish a frequency.
Normative research questions seek to establish what is actually happening in contrast to the norm or what should be happening. An example of a normative question is “to what extent are private universities complying with health and safety laws?” Normative questions may be answered with a narrative rather than the use of a hypothesis.
Correlative questions seek to establish relationships and are interested in the strength of relationships between variables. An example of a correlative question is “what is the relationship between management style, university ranking and compliance with health and safety regulations?” While correlative questions may be tested by hypotheses, they do not seek to establish causality but intend to ‘correlate’ variables and show the existence or otherwise of significant relationships between them.
Impact questions seek to establish causality in relationships between variables and measure the effect which variables have on one another (i.e. what impact does a change in variable x have on variable y?). An example of an impact question is “does compliance with health and safety regulations have an impact on productivity?” Impact questions may also be tested by hypotheses.
Examples of Research Questions
Research questions are derived from the objectives of the study and should help to break down the problem of study into manageable units. If the problem of the study is about the relationship between input metrics and the effectiveness of universal basic education, where input metrics are defined by:
The teacher-pupil ratio
Availability of audio visual tools
Availability of learning materials including chalkboards, chairs and desks, and stationery
Valid research questions in this scenario would include:
Does universal basic education receive critical support from the government regarding its input metrics?
What is the relationship between the teacher pupil ratio and the effectiveness of universal basic education?
What is the impact of classroom availability on the effectiveness of universal basic education?
What is the relationship between textbook availability and the effectiveness of universal basic education?
Is the effectiveness of universal basic education influenced by the availability of audio visual tools?
What is the relationship between availability of learning materials and the effectiveness of universal basic education?
The first question does not require the use of a hypothesis but the other questions do. This is because the first question can be approached by using a narrative based on qualitative study using observation and primary and secondary data.
These different variables, as mentioned earlier, should each have indicators which are measurable and attainable in order to properly use the research questions above.
Linking Research Questions to Other Parts of the Study
Research questions are derived from the purpose and objectives of the study and are often tied to each hypothesis, which are logical extensions of the research questions themselves. Research questions help to narrow the focus of the research on researchable areas within the scope of the study.
The research questions can be directly linked with the findings of the research in qualitative studies where there are no hypotheses. In this case, questions may be used to guide the researcher and when a tangential point becomes of greater interest, new questions are used to determine the ways in which to approach the research bearing in mind the new evidence uncovered.
The research questions are also linked directly to the research problem which underpins the study. The research problem may in itself be too wide in scope to be satisfactorily tackled by limited research. The research questions therefore, help to streamline the research into areas which may be studied.
Limitations of Research Questions
Research questions are useful inasmuch as they ask questions which can be linked directly with the objectives of the study. In quantitative studies, just like hypotheses, they help to narrow down the area of focus for the research on acquiring information which is researchable and desirable. In qualitative studies they help to set the focus for commencing discussions into the area of study in order that data may be generated.
Research questions are simply statements of a focused objective of the study. They are used especially when the researcher may not want to specify the direction of the relationship among the variables, or they may be used to determine the key variables from the research objectives.
Comparing and Contrasting Hypotheses with Research Questions
The following is a point by point contrast and comparison between hypotheses and research questions
Generated from qualitative research but used in quantitative studies
Used in both quantitative and qualitative research
In a hypothesis, the predicted relationship between variables is either true or false
With a research question, the answer to the question may be closed (e.g. a yes or no answer) or may be open and descriptive in nature.
A hypothesis is empirically verifiable
The answer to a research question may not always be verifiable
Hypotheses are generated from research questions
Research questions are generated from the research problem. They may or may not generate a hypothesis
The hypothesis helps to test a relationship between variables i.e. can be tested and verified
The research question seeks to help gather preliminary data for a study
In Qualitative research
Hypotheses are generated
Research questions are used as a guide
As can be seen from the above, research questions are quite integral to the concept of research just as much as they are indelibly linked to hypotheses, in quantitative research. The two are used as an aid to delimiting concepts for empirical study in quantitative research and while research questions are used to guide qualitative research, hypotheses are products of qualitative research.
Summary and Conclusion
The focus of this paper has been to show the various differences between hypotheses and research questions and how these concepts are used in research. These have been highlighted starting with functional definitions, explanations and limitations of the various concepts involved. The different types of hypotheses and research questions have also been enumerated and explained while showing how they may be used in research.
The paper has shown concrete contextual examples to illustrate the different concepts from a local perspective and examples of each concept have been used to aid the reader at each step. This paper has concluded with a tabular exposition of the contrast between research questions and hypotheses for validation of premises earlier outlined.