Quantitative Research
Introduction[edit|edit source]
Quantitative research is atype of researchthat assumes that the phenomena under study can be measured and involves methods that:
- gather data using measurement (numerical data)
- analyse data by using quantitative statistical analysis techniques.
Quantitative Research Methodology[edit|edit source]
Methodology refers to the overall approach taken in a piece of research.[2]In quantitative research, it encompasses the general principles of investigation that guide how such a study is designed and conducted in order to answer a quantitative research question.
The methodological steps for quantitative investigations are found below.
Step 1: The Research Question[edit|edit source]
任何研究的核心是研究探索ion(s).[2]Research question(s) guide the design and methods used in a study, being key to not only identifying gaps in knowledge but also refining and adjusting existing knowledge.[2]
In quantitative research, the nature of the research question may be descriptive / normative or explanatory.
Descriptive / normative research questions[edit|edit source]
Descriptive research questions provide a descriptive account of a phenomenon within an established framework of knowledge; this approach is often used when aiming to develop a fuller account of an observation and is sometimes combined with identifying some relationships of potential interest.[2]For example, a researcher may choose to utilise survey methods to investigate the characteristics, perceptions and behavious about a particular condition in a specific population.[3]
Normative research questions are similar in their purpose to descriptive research questions, but also include an additional objective of comparing data gathered with a criterion or standard.[4]
Explanatory research questions[edit|edit source]
Explanatory research questions usually test a hypothesis i.e. a prediction that the study sets out to either retain or reject, by means of statistical inference testing.[5][6]
Step 2: Research strategy[edit|edit source]
Quantitative investigations may be primary or secondary, depending on the source of research knowledge.
Primary research[edit|edit source]
Primary investigations involve the actual research study i.e. information gathered through self-conducted research methods. The decision on primary research is influenced by the number / quality of available information. A search for available evidence is common before any research study. Primary research study is usually based on gaps in available knowledge.
Secondary research[edit|edit source]
Secondary research accesses primary data through previously conducted quantitative research studies. Systematic reviews and meta-analyses that gather and analyse clinical/experimental primary studies rank higher than single quantitative studies in thehierarchy of evidence.[7]
Step 3: Designs[edit|edit source]
Primary research[edit|edit source]
There are five main types of primary quantitative research designs:[1]
- Descriptive
- Survey
- Correlational
- Quasi-experimental
- Experimental
Descriptive research designs[edit|edit source]
As the name implies, descriptive research aims to observe and measure a phenomenon e.g. identify characteristics, categories or describe a patient’s journey e.g. case study.[8]
Survey research designs[edit|edit source]
Surveydesigns are most frequently employed in healthcare epidemiology research.[9]Surveys may be used to gain insights into opinions and practices in large samples; they can be descriptive and/or be used to test associations.[9]
Cross-sectional survey researchstudies exposures and outcomes at a certain point in time.[10]These observational designs are commonly used for population-based studies and investigations of disease prevalence.[10]
Longitudinal survey researchmeasures at different points in time (continuously or repeatedly).[11]This design is very useful in identifying the relationship between risk factors and disease onset, or treatment outcomes over time.[11]
Correlational research designs[edit|edit source]
Correlational research designs identify relationships between variables without implying causation. This can be done in terms of direction and/or strength of the relationship between two variables with no influence of any extraneous variable, or manipulation.[12]
Cohort studiesare prospective or retrospective in nature. A sample of participants is observed over time where those exposed and those not exposed are compared for differences in one or more predefined outcomes, such as adverse event rates.[12]
Case-control studiesare retrospective in nature, participants already exposed to the event are selected, then matched with unexposed participants, using historical cases to ensure they have similar characteristics.[12]
Cross-sectional studiesare a type of cohort study where only one comparison is made between exposed and unexposed subjects.[12]
Quasi-experimental research designs[edit|edit source]
Aims to identify a cause-and-effect relationship between independent and dependent variables but based on no random criteria. Commonly found as non-randomised, pre-post intervention studies.[13]
Experimental research designs[edit|edit source]
In experimental research designs, the researcher can manipulate one (or more) variable(s), the independent variable, and study the effect on a dependent variable.[1]There are many types of experimental designs; one of the most important is the randomised controlled trial.
Randomised controlled trialsare considered top methods in thehierarchy of evidencewhen testing the link betweencause and effectin clinical interventions.[14][15]
Secondary research[edit|edit source]
Meta-analyses[edit|edit source]
Meta-analyses synthesise results from multiple studies in a quantitative manner to determine the average effectiveness of interventions.[16]
Step 4: Data analysis[edit|edit source]
Data analysis in quantitative research involves the use of statistics to investigate numerical data.[17]This may involve descriptive statistics for distributions and relationships between variables and inferential statistics.[17]
Step 5: Data reporting[edit|edit source]
Reporting and interpreting quantitative research findings usually follows a set of guidelines.[18]Recommendations may involve the reporting of designs, methods and procedures, data and analyses, usually with a aim of producing a publishable result.[19]
Key differences between different forms of quantitative research[edit|edit source]
Key differences between different forms of quantitative research are found below. Distinctions have been simplified for clarity and some overlap may exist between characteristics.[2][6]
Descriptive / Normative | Explanatory | |
---|---|---|
Ontological assumption | Reality is objective and singular apart from the researcher. | Reality is objective and singular apart from the researcher. |
Nature of research question | Fairly specific and largely definite | Highly specific and definite; Declarative |
Research design | Structured, sequential and largely predetermined | Highly structured, sequential and predetermined |
Data collected | Quantitative | Quantitative |
Relationship to theory | Aims to develop or elaborate theory | Hypothesis testing |
References[edit|edit source]
- ↑1.01.11.2Watson R.Quantitative research. Nurs Stand. 2015 Apr 1;29(31):44-8.
- ↑2.02.12.22.32.4Sim J., Wright C. Research in Health Care: concepts, designs and methods. Nelson Thornes: Cheltenham, UK. 2002
- ↑Mouchtouri V., Agathagelidou E., Kofonikolas K., Rousou X., Dadouli K., Pinaka O., et al.Nationwide Survey in Greece about Knowledge, Risk Perceptions, and Preventive Behaviors for COVID-19 during the General Lockdown in April 2020.Int J Environ Res Public Health. 2020 Nov 28;17(23):8854.
- ↑Hedrick T., Bickman L., Rog D. Applied Research Design: A Practical Guide. Sage Publications, Newbury Park. 1993
- ↑Hazra A., Gogtay N.生物统计学系列模块2:Hypothe的概述sis Testing. Indian J Dermatol. 2016 Mar-Apr;61(2):137-45.
- ↑6.06.1Sukamolson S. Fundamentals of quantitative research. Available:https://www.researchgate.net/profile/Vihan-Moodi/post/What_are_the_characteristics_of_quantitative_research/attachment/5f3091d0ed60840001c62a27/AS%3A922776944787456%401597018576221/download/SuphatSukamolson.pdf(accessed 27/3/2023)
- ↑Kapoor M.Types of studies and research design. Indian J Anaesth. 2016 Sep; 60(9): 626–630.
- ↑Crowe S., Cresswell K., Robertson, A., Huby G., Avery A., Sheikh A.The case study approach.BMC Med Res Methodol 2011,11:100.
- ↑9.09.1Safdar N., Abbo L., Knobloch M., Seo S.Research Methods in Healthcare Epidemiology: Survey and Qualitative Research.Infect Control Hosp Epidemiol. 2016 Nov; 37(11): 1272–1277.
- ↑10.010.1Setia MS.Methodology Series Module 3: Cross-sectional Studies.Indian J Dermatol. 2016 May- Jun;61(3):261-4.
- ↑11.011.1Caruana E., Roman M., Hernández-Sánchez J., Solli P.Longitudinal studies.J Thorac Dis. 2015 Nov;7(11):E537-40.
- ↑12.012.112.212.3Lau F. Chapter 12: Methods for correlational studies. Handbook of eHealth evaluation: an evidence-based approach. Available:https://www.ncbi.nlm.nih.gov/books/NBK481614/(Accessed 26/03/2023)
- ↑Harris A., McGregor J., Perencevich E., Furuno J., Zhu J., Peterson D., Finkelstein J.The use and interpretation of quasi-experimental studies in medical informatics.J Am Med Inform Assoc. 2006 Jan-Feb;13(1):16-23.
- ↑Evans D.Hierarchy of evidence: a framework for ranking evidence evaluating healthcare interventions.J Clin Nurs. 2003 Jan;12(1):77-84.
- ↑Wallace S., Barak G., Truong G., Parker M.Hierarchy of Evidence Within the Medical Literature.Hosp Pediatr. 2022 Aug 1;12(8):745-750.
- ↑Hernandez AV, Marti KM, Roman YM.Meta-Analysis. Chest. 2020 Jul;158(1S):S97-102
- ↑17.017.1Sheard J.Chapter 18 - Quantitative data analysis. Research methods (second edition) 2018; 429-52.
- ↑Larson-Hall J., & Plonsky L.Reporting and Interpreting Quantitative Research Findings: What Gets Reported and Recommendations for the Field.Language Learning 2015; 65(S1): 127-159.
- ↑Norris J., Plonsky L., Ross S., Schoonen R.Guidelines for Reporting Quantitative Methods and Results in Primary Research.Language Learning 2015; 65(2):470-476.