If we compare research to setting out on a journey, then the research question is the statement of the destination. We cannot map a clear and effective route unless we know where we are going, and we certainly will not know whether we have arrived, unless we know where we wanted to be at the end of the journey. In the same way, the research question allows the researcher to plan the research in the best possible way, and make important decisions to ensure that the correct destination is reached.

The following are aspects of the research process that will be influenced by the research question:

We can see now what Atkinson (2008: 67) had in mind when he said:

Research questions evolve from the choice of a particular topic area. This is often a topic felt to be problematic or where questions are raised on what is best practice for optimum care. The choice of topic can emerge from a desire to improve the quality of services, whilst others arise from reviewing the literature, or from searching for ways to provide clinically effective care. In fact, according to Houser (2008), the search for research problems is one of the easiest parts of the research process, and that is certainly true within midwifery.

Can research answer every kind of question? The answer is ‘no’. Some questions demand a value judgement for their resolution, and are not open to research. For example, ‘Should midwives carry out some of the more technological procedures currently performed by obstetricians?’ Although we can survey midwives’ views or those of obstetricians, the answers would not indicate whether it is ‘right’, only what people feel about it. Similarly, some questions are ethical or philosophical questions and cannot be answered by research but need to be discussed and debated.

Further light is shed on this issue by Burns and Grove (2009: 76) who propose that questions about practice fall into three categories:

The first option is important as it illustrates that we should not rush into research without checking whether there is already knowledge available to answer our question. That is why a review of the literature is such an important preliminary stage to starting research. Their second category relates to the need for debate, reflection and problem solving. The third category is the one in which we are interested.

One important consideration before pursuing a research question is that of relevance. Does the research need to be done? Every research idea should be evaluated in terms of the contribution it will make to midwifery: ‘Do we need to know the answer to the question?’ This can be characterised as the ‘so what’ and ‘who cares’ test, which ensures that thought it given to the outcome of research and its relevance to practice, the development of theory or its contribution to shaping policy.

Perhaps the most important criteria in judging the relevance of a research project is whether women, their babies and their family gain from this research? Even if the topic relates to midwives themselves, those receiving care may still benefit indirectly through an increase or change in midwives’ knowledge, skills or attitudes.

Having considered the relevance of a particular study, the next issue is that of feasibility. This includes such factors as:

The researcher must be able to confirm that all of these issues can be successfully addressed before the research can go any further.

Research questions are structured in a number of different ways according to the level of question (Wood and Ross-Kerr 2006) (see Chapter 2). The way the question is written or ‘framed’ will illustrate the level it addresses. Each level is associated with an appropriate broad research approach. For example, a level-one question will suggest the use of a survey or a qualitative approach such as an ethnographic or phenomenological study. Level-one questions are those where little is known about a topic and the intention is to describe a situation. There is only one variable in a level-one question. The researcher should give a clear concept definition that relates to the way the variable will be defined for the purposes of the study. There should also be an operational definition in a quantitative study that will outline the way it is intended to measure that variable. At this level there is no attempt to establish cause-and-effect relationships between variables.

A level-two question may also suggest a survey, but the question will be concerned with the pattern or correlation between variables. This level may also involve the collection of physiological measurements through observation or taking samples where at least two different measures from each subject are compared statistically to see if they show a similar pattern or correlation. In a level-two question more is known about the topic. Here, the purpose of the research is to establish if there is a statistical relationship in the form of a correlation between the variables that have been identified. At this level, according to Wood and Ross-Kerr (2006), although the researcher might have a shrewd idea of what to expect, there is not enough firm evidence from a randomised control trial to confidently predict an outcome and so achieve a level-three question.

A level-three question will look for a cause-and-effect relationship between two variables, particularly in relation to a clinical outcome. In a level-three question there will be enough known about the nature of the relationship between the variables in the study to make a confident prediction about relationships and outcomes. The purpose is to examine why a relationship exists, or to test a theory. This is achieved by manipulating the independent variable to measure its effect on the dependent variable in an experimental design study such as a randomised control trial. Some of the questions midwifery research attempts to answer and how these are related to the level of question are shown in Table 7.1.