Vous êtes sur la page 1sur 5


During the late 19th and throughout the 20th century, strategies of inquiry associated with quantitative research were those that invoked the postpositivist worldview and that originated mainly in psychology. These include true experiments and the less rigorous experiments called quasiexperiments (see, an original, early treatise on this, Campbell & Stanley, 1963). An additional experimental design is applied behavioral analysis or single-subject experiments in which an experimental treatment is administered over time to a single individual or a small number of individuals (Cooper, Heron, &Heward, 2007; Neuman& McCormick, 1995).

One type of nonexperimental quantitative research is causal-comparative research in which the investigator compares two or more groups in terms of a cause (or independent variable) that has already happened. Another nonexperimental form of research is the correlational design in which investigators use the correlational statistic to describe and measure the degree or association (or relationship) between two or more variables or sets of scores (Creswell, 2012). These designs have been elaborated into more complex relationships among variables found in techniques of structural equation modeling, hierarchical linear modeling, and logistic regression. More recently, quantitative strategies have involved complex experiments with many variables and treatments (e.g., factorial designs and repeated measure designs). They have also included elaborate structural equation models that incorporate causal paths and the identification of the collective strength of multiple variables.

QUANTITATIVE DESIGNS During the late 19th and throughout the 20th century, strategies of inquiry associated with

Survey research provides a quantitative or numeric description of trends, attitudes, or opinions of a population by studying a sample of that population. It includes cross-sectional and longitudinal studies using questionnaires or structured interviews for data

collection—with the intent of generalizing from a sample to a population (Fowler, 2008).

Experimental research seeks to determine if a specific treatment influences an outcome. The researcher assesses this by providing a specific treatment to one group and withholding it from another and then determining how both groups scored on an outcome. Experiments include true experiments, with the random assignment of subjects to treatment conditions, and quasi-experiments that use nonrandomized assignments (Keppel, 1991). Included within quasi-experiments are singlesubject designs.

Reference: Cresswell, John Research Design:

Qualitative, Quantitative, and Mixed Methods Approaches


Quasi-experimental research may look very much like true experimental research in that it does involve the manipulation of an independent variable.

However, it is not the same as true experimental research because quasi-experimental research studies lack one or both of the essential properties of randomisation and a control group.




with quasi-

experimental research is that, compared


experimental research, it has a weakness in that is not

possible to deliver 'cause and effect' results.


other words, we cannot infer from quasi-

experimental research that, for example, doing one

thing causes a particular phenomenon (e.g. smoking cigarettes causes cancer).


This type of research describes what exists and may help to uncover new facts and meaning. The purpose of descriptive research is to observe describe document






(Polit&Hungler 1999)




This involves the collection of data that will provide an account or description of individuals, groups or situations. Instruments we use to obtain data in descriptive studies include

As pointed out in the accompanying book (chapter 5), depending upon the problem being investigated, this type of research may be better explored by means of a qualitative paradigm.


interviews (closed questions)


observation (checklists, etc.)

There is no experimental manipulation or indeed any

random selection to experimental research.






The characteristics of individuals and groups such as nurses, patients and families may be the focus of descriptive research. It can provide a knowledge base which can act as a springboard for other types of quantitative research methods.


Quantitative correlational research aims to systematically investigate and explain the nature of the relationship between variables in the real world. Often the quantifiable data (i.e. data that we can quantify or count) from descriptive studies are frequently analysed in this way.

Correlational research studies go beyond simply describing what exists and are concerned with systematically investigating relationships between two or more variables of interest (Porter & Carter


Such studies only describe and attempt to explain the nature of relationships that exist, and do not examine causality (i.e. whether one variable causes the other).


According to Polit and Hungler (1999), a survey is used to obtain information from groups of people (i.e. populations).

The information that is obtained may be concerned with the prevalence, the distribution, and/or the interrelationships between variables within these groups.

Data collection tools include:

personal interviews

telephone interviews


This is an “applied form of research that involved finding out how well a programme, practice, procedure or policy is working” (Polit&Hungler 1999:201).The aim of this type of research is to assess/evaluate the success of a particular practice or policy, etc.

Examples of this type of research can be seen in various types of analysis/evaluation, including:

process/implementation analysis (the analysis of process and the implementation of treatments/nursing cares, etc. - similar to audits)

outcome analysis (the analysis of the outcome of changes in processes, treatments, etc. - again, similar to audits)

impact analysis (the analysis of the impact that, for example, a new treatment will have on the patients)

cost-benefit analysis (the analysis of the cost to benefit ratio of , again for example, the introduction of a new drug)

As with the survey research methods, this type of research may best be carried out as a qualitative piece or research, depending upon the original research question.


Cresswell, John Research Design: Qualitative, Quantitative, and Mixed Methods Approaches

Polit DF, Hungler BP (1999) Nursing Research:

Principles and Methods (6th Ed.) Philadelphia, Lippincott

Porter, S., Carter, DE (2000) Common terms and concepts in research. In Cormack, D. (Ed.) The Research Process in Nursing (4th Ed.). Oxford, Blackwell Science (pp. 17-28)


There are a variety of techniques that can be used to

collect data in a quantitative research study. However, all of them are geared towards numerical collection. This numerical data can be collected by means of:

In quantitative research, the data are collected and recorded systematically, and these are then organised

so that they can be entered into a computer database.


In quantitative research, the observation must be structured so that there is a defined purpose to the observation. The first step in structured observational measurement is to very carefully define what is to be observed. Once the decision has been made as to what is to be observed, the next step is to make decisions as to how the observations are to be:



coded. In most cases, a category system is developed for organising and sorting the behaviour or events that are being observed. The categories that are to be observed should be mutually exclusive.

The observer may use checklists as an aid to the structured observation. quantitative data collection


Although interviews are usually associated with qualitative research, they can have a role to play in quantitative research as well. In the case of quantitative research, the interview will be totally structured, with the interviewee only being able to choose a response (usually one word) from a series on the interview form.

Often the reply can only be 'yes' or 'no', or it may just be a number. Alternatively, the interviewee may be asked just to choose one item from a list. These replies can then be coded and entered into a computer database for statistical analysis. The interview may well be linked to a checklist. An interview is used often in research studies where there is a poor return rate of postal questionnaires and checklists.


Rating scales can be used for observation as well as

self-reporting. A rating scale allows the observer to rate the behaviour or event on a scale. Rating scales are the crudest form of measure involving scales. A rating scale lists an ordered series of categories of a variable that are assumed to be based on an

underlying continuum. A numerical value is then assigned to each category in each item. This type of scale is often used in observational

measurement to guide data collection. There is, however, another type of scale that you will often see in research papers - the Likert Scale.


Much of nursing practice is linked to physiological

dimensions of health. Physiological measurements can be obtained by using a variety of methods, namely:




direct measurement;

indirect measurement;

laboratory tests;

electronic monitoring;

other creative methods of obtaining these


Questionnaires can be useful for collecting data on simple and well-defined issues. Their design should be carefully planned and piloted to ensure that they provide:

the required data;

data that can be analysed and used;

an unbiased response.

There are two types of questions that you can include in a questionnaire, depending particularly upon whether it is going to be a quantitative or a qualitative questionnaire. These are:

closed questions;

open-ended questions. Often, we are warned against 'leading questions' in questionnaires - however, all questions are 'leading'. Some questions, however, lead more than others; but they all lead towards an answer. Constructing a reliable and valid questionnaire to collect high- quality data is a subtle and sophisticated art. Poorly designed questionnaires collect poor quality data.

Instrument is the generic term that researchers use for a measurement device (survey, test, questionnaire, etc.). To help distinguish between instrument and instrumentation, consider that the instrument is the device and instrumentation is the course of action (the process of developing, testing, and using the device).

Instruments fall into two broad categories, researcher- completed and subject-completed, distinguished by those instruments that researchers administer versus those that are completed by participants. Researchers chose which type of instrument, or instruments, to use based on the research question. Examples are listed below:

Researcher-completed Instruments


completed Instruments Questionnaires

Rating scales Interview schedules/guides

Self-checklists Attitude scale Personality inventories

Tally sheets Flowcharts Performance checklists

Achievement/aptitude tests

Time-and-motion logs

Projective devices Observation forms Sociometric devices


Writing the methodology lies at the core of the paper, and fulfills one of the basic principles underlying the scientific method.

Any scientific paper needs to be verifiable by other researchers, so that they can review the results by replicating the experiment and guaranteeing the validity.

To assist this, you need to give a completely accurate description of the equipment and the techniques used for gathering the data.

Finally, you must provide an explanation of how the raw data was compiled and analyzed.

Writing Methodology Allows Verification

Other scientists are not going to take your word for it, and they want to be able to evaluate whether your methodology is sound.

In addition, it is useful for the reader to understand how you obtained your data, because it allows them to evaluate the quality of the results.

For example, if you were trying to obtain data about shopping preferences, you will obtain different results from a multiple-choice questionnaire than from a series of open interviews.

Writing methodology allows the reader to make their own decision about the validity of the data.

If the research about shopping preferences were built upon a single case study, it would have little external validity, and the reader would treat the results with the contempt that they deserve.

The Structure Behind

Whilst there are slightly different variations according to the exact type of research, the methodology can be divided into a few sections.

Describe the materials and equipment used in the research.

Explain how the samples were gathered, any randomization techniques and how the samples were prepared.

Explain how the measurements were made and what calculations were performed upon the raw data.

Describe the statistical techniques used upon the data.

That is the very basic structure of writing methodology, and it will clarify all of the information.

The writing for the method should be clear and direct, concise and straight to the point. The major point is not to stray off into irrelevance, and this process is helped by making a few basic assumptions.

For example, in a psychology paper, there is no need to describe a Skinner box, as that is well known to psychologists. However, you would need to explain exactly how the box was used, to allow exact replication.

Whilst not always possible, the methodology should be written in chronological order, always using the past tense. Writing Methodology at the Core of the Research Paper

A well laid out and logical methodology will provide a great backbone for the entire research paper, and will allow you to build an extremely strong results section.

The only real difficulty with the methods section is finding the balance between keeping the section short, whilst including all of the relevant information.

The other problem is finding the correct style of writing:

APA guidelines suggest that you should use 'I' and 'We', but most supervisors still prefer an impersonal passive tense. Check this with your supervisor before you start writing, to avoid unnecessary editing!