“If I go out into nature, into the unknown, to the fringes of knowledge, everything seems mixed up and contradictory, illogical and incoherent. This is what research does; it smoothes out contradiction and makes things simple, logical and coherent.” (Waltz et al 1991) Research delivers information. The more information we have about the population, the more we are able to influence and control the health services. Research may be descriptive, dealing with questions like “how many”, “who”, “what’, or it may be explanatory, looking at “why”. How is this information gained? There are a variety of methods available, however none have been found to be perfect. Hewison (1995) argues that research is done to help health professionals make better decisions. Practice, care, and the service they provide can only ever be as good as the knowledge on which it is based.
Meanwhile, knowledge derived from research has two main advantages: its relative objectivity, and its capacity to handle complexity. Objectivity is an attempt to apply the same standards of evidence to things they do and do not believe in. Furthermore, familiarity with research methods might reduce the likelihood of jumping to conclusions. Complexity is the other reason why research needs to be carried out as a guide to good practice.
The different methods of research techniques enable policy makers to reach better and more informed decisions. This leads on to the ‘how’ of health services research. The ‘how’ is a set of methods (hard and soft data within this context) which enable health professionals to detect patterns and to have confidence that the interpretation that is placed on them is the correct one (Hewison 1995).
Hard Data Couchman and Dawson (1990) compare quantitative data to ‘hard’ data generated from standardized questionnaires, while qualitative data represents ‘soft’ data derived from indepth interviews. They (Couchman and Dawson 1990) further classify these differences as producing inductive logic through qualitative methods from one extreme, to deductive logic through quantitative methods on the other. It may be argued that hard and soft data are terms used to describe data derived from a positivistic and an antipositivistic approach respectively.
“Over the past three decades, the majority of nurse researchers have been strongly socialized to value and use quantitative types of research as the only legitimate form of ‘scientific’ nursing approach.” (Leininger 1985) The scientific collection of data by methods of the natural sciences is known as positivism. The collection of statistical information, which is “free of bias”, could enable researchers to uncover the deeper courses of human behavior.
The object of research according to Leininger (1985) is to develop social policy based on accurate information in order to tackle social problems and to improve the quality of life for the average citizen. However the quotation above reflects the tendencies of nurse researchers to abide to statistical data that seems to generate an atmosphere of certainty of how to apply the knowledge in practice. Smeltzer and Hinshaw (1993) when referring to nursing research argue that: “In nursing administration there is a critical need for accurate information on which to base both clinical and administrative policy.”
It may be argued that the authors (Smeltzer and Hinshaw 1993) are comparing ‘accurate information’ to ‘hard data’. Even the word ‘hard’ is indicative as something concrete, black on white, statistically clear. Hunt (1993) emphasizes that the literature on the research process for nurses has reflected these positions stated above and concentrate largely on methods represented as “rational, objective, and quantitative.” Hunt (1993) also confirms that nursing literature has tended to consider only quantitative research producing numerical data and with widely applicable results to be ‘scientific’ and of a high status.
In the authors’ view, the advantages of ‘hard’ data in the context of decision making in the health services is that it may generate a tendency of acceptance through objective analysis, free from biases and subjectivity. Quantitative research usually involves large numbers of respondents in tightly structured investigations where the primary concern of the researcher is to establish incidence and to ascertain patterns which indicate structural regularities (Stanley and Wise 1990). However the quantitative research debate raises the fundamental question and asks: “Can human beings and their social endeavours be studied in the same way as rats, plants, and planets?” (Hunt 1993)
Hard data obtained from surveys can be problematic. Even if a questionnaire design is sound, and the questions asked are clear and appropriate, the answers given may not necessarily be accurate (Stanley and Wise 1990). Sheehan (1986) admits that nursing research is perhaps facing the biggest challenge; that of applying it to nursing practice. According to Bond (1993), it is always advantageous therefore to evaluate researches in order to provide relevant information for decision makers to set priorities, allocate the necessary resources, and to modify and refine project structures and processes.
He (Bond 1993) states that a major criterion (in the context of decision making) in considering whether the evaluator has done a good job appears to lie in the truth of the findings. Can the researcher be believed regardless of any other value the research may hold? In order to validate evaluative research, it appears to be necessary to conform to strict principles of science and measurement. Bond (1993) strengthens the argument and confirms that: “Like most of the social science, early evaluation looked to logical positivism to justify its method choices. Congruent with this approach was the performance for using goals articulated in advance as a basis for formulating causal hypotheses which could then be tested experimentally.”
Soft Data
Qualitative methods facilitate the study of selected issues in depth or in detail. Qualitative research can be naturalistic, that is, it can take place in a setting in which the respondent is comfortable at home. Events which then occur are natural in the sense that they are not planned or constructed by the social scientist. This kind of research is often referred to as fieldwork. This encompasses becoming involved in the social interaction or group in a way that prevents detachment. Objectivity of the researcher is highly questioned. However, Hewson (1995) states that closeness does not automatically indicate bias, and distance is no guarantee of detachment.
Also it can be argued that without empathy there cannot be comprehensive understanding. In the absence of pre-determined categories into which responses must be fitted, a more meaningful and valid account can emerge. Analysis can be more difficult because responses are not systematic or standardized and are difficult to categorize. Yet this approach can allow the researcher to see the world as the respondent sees it. It is highly labour intensive, and therefore an expensive self limiting strategy (Polit and Hungler 1995).
Qualitative work is essential to the knowledge development of the health care disciplines. If credence is given to the creativity and intellectual agility required in qualitative research then this does not allow for methodological or intellectual slappinness. Intuition, creativity and intellectual agility are most often found within the context of careful and vigorous attention to method (Burns and Grove 1993).
Only until recently there has been awareness and debate for the “process involved in knowledge creation or the different philosophical positions that underpin the choice of qualitative research methods” (Hunt 1993). Although qualitative research is expressive and time consuming, its comprehensiveness enables decisions to be made about the most appropriate policy to adopt and also an understanding of where the policy may be most successfully implemented. Gortner (1984) cited by Treece and Treece (1986) call for greater explanatory power in nursing research as a means of influencing policy formation.
Such explanatory power also gains the respect of researchers in other disciplines as they recognize the contribution that nursing qualitative research makes to new knowledge. The expansion in understanding the human being – the focus on all nursing – cannot be comprehended at the present status of nursing research. But as nursing investigators develop new modes of inquiry through description and induction as well as prescription and deduction, the answers to why and the meaning will result.
According to Pollock (1993) unless researchers show clearly the methods they use in order to manage the large amounts of qualitative data collected, the advancements of knowledge within the social sciences and the conclusions drawn are not considered reliable. If available tools to manage data are not utilised, analysis and presentation of findings may not take place. Pollock (1993) admits though that “Qualitative analysis is not easy and the findings of such studies can be challenged on the grounds of questionable reliability and validity”.
Discussion
Hewson (1995) cites the Medical Research Council and the Welcome Trust who both have different views towards health services research. The medical Research Council defines health services research as: “the identification of the health and needs of a community and the study of the provision, effectiveness and use of the health services.” On the other hand the Welcome Trust has referred to it as: “…the identification and quantification of health care needs and the quantitative study of the provision and use of health services to meet them.” Naturally not everybody would agree with the emphasis on quantification apparent in the second definition.
The qualitative or soft approach in research relies on concentrated observation and judgment (Hewison 1995). For example, data is collected in an unstructured way through free-ranging interviews which will then be the source of information. According to Reid (1989) the information collected and the conclusions reached will rarely be generalized beyond the immediate context. Conversely in the quantitative or ‘hard’ research approach, well defined hypotheses are generated, results will often be generalised, and their conclusions are based solely on empirical evidence (Reid 1989).