Quantitative Research Notes

INTRODUCTION TO QUANTITATIVE RESEARCH

• Quantitative research is a formal, objective, and systematic process. It aims to:
  • Describe and test relationships.
  • Examine cause-and-effect interactions among variables.
• Quantitative studies require a research design that allows consumers to understand:
  • The research questions.
  • How the study was conducted.
  • What the results indicate.

RESEARCH DESIGN

• Research design is a blueprint for conducting the study that:
  • Maximizes control over factors that could interfere with the validity of the findings.
• Design encompasses:
  • The plan.
  • The strategy.
  • The structure for hypothesis testing or answering research questions.
• The researcher must:
  • Clearly describe how the study will be conducted before starting the project.
• This enables the researcher to:
  • Apply different levels of control.
  • Suggest that the independent variable influenced the dependent variable, and not something else.

QUANTITATIVE PROCESS

• Research design is influenced by:
  • Problem statement.
  • Hypothesis.
  • Theoretical framework.
  • Literature review.

CONCEPTUALIZATION OF THE PROBLEM

• Conceptualization of the research problem is derived from:
  • The literature review.
  • The development of the theoretical framework.
• Literature review considerations:
  • When was the problem studied?
  • Which aspects of the problem have been studied?
  • Where was the problem investigated?
  • Who has investigated the problem?
  • Are there gaps or inconsistencies in the literature?
• The literature review informs us:
  • What we currently know about the research problem.

THEORETICAL FRAMEWORK

• The researcher will examine the theoretical framework, which is:
  • The structure or meanings that guide the development of the study.
  • A description of how this will link to nursing’s body of knowledge.
• Key questions to consider:
  • How are the variables connected?
  • How does this connect with nursing?
• Example: Kolb’s Learning Cycle
  • Abstract – lecture about blood pressure.
  • Active – Lab: practicing taking a blood pressure.
  • Concrete – taking blood pressure at clinical sites on patients.
  • Reflection – reviewing results with faculty.

ELEMENTS OF THE DESIGN

• Elements of the Design:
  • Participants (who).
  • Observations (what).
  • Measurement of time (when).
  • Selection of participants (where).
  • Role of investigator.

RESEARCH DESIGN CONSIDERATIONS

• Research Considerations – must be taken into account before the study is conducted:
  • Objectivity.
  • Accuracy.
  • Feasibility.
  • Control.
    • Homogeneous sample.
    • Constancy.
    • Manipulation.
    • Randomization.

RESEARCH DESIGN CONSIDERATIONS - DETAILS

• Objectivity:
  • Review of the literature, assessing what is known about the research problem(s).
• Accuracy:
  • Using a review of literature helps to determine the best theoretical framework that will accurately, logically, and systematically examine the research problem.
• Feasibility:
  • The ability of the study to be successful.
  • Examine the cost of the study.
  • Availability of the participants.
  • Timing.
  • Data analysis processes.

PILOT STUDY

• Pilot study:
  • A simple small study that is conducted to determine the objectively, accuracy, and feasibility of the study before embarking on an extensive experimental study.
• It examines:
  • Participant availability.
  • The intervention.
  • Data collection protocols.
  • Likelihood participants will complete the study.
  • Reliability and validity of measuring tools.
  • The cost of study.

RESEARCH DESIGN - CONTROL

• Control:
  • Measures that the researcher uses to keep the study uniform (consistent) and prevent bias.
  • Maximizes the degree of control over testable variables.
• Example: How a new smoking cessation program (IV) affects smoking behaviors (DV).
  • The design needs to ensure that the new cessation program impacts smoking behaviors.
  • Smoking behavior is not impacted by other factors such as motivation, medications, family support, or peer pressure – Extraneous (mediating) variables.
• Bias:
  • Any action or influence that distorts the finding of the study.

STUDY CONTROLS

• To prevent an extraneous variable from interfering with the operation of the phenomena being studied, controls are put in place.
• Homogenous sampling:
  • Having a very similar sample.
  • Characteristics that can be controlled – age, gender, medical history as relevant = inclusion criteria.
  • Example if studying community support for Canadians over 65 years of age in Atlantic Canada.
    • Inclusion criteria: over 65 years of age, living in Atlantic Canada, Canadian citizenship.
  • This can limit the generalizability of the study – who the study results can be applied to.

STUDY CONTROLS - CONTINUED

• Consistency in Data Collection:
  • Each participant is exposed to the same environmental conditions, timing of data collection, data collecting instruments, and data collection procedures.
  • Example: The nurse collects the study’s participants’ BP one hour after meditation therapy by trained personnel in the same setting (Clinic) using the same BP cuffs (that were calibrated at the beginning of the day).
  • This consistency allows the researcher to draw clear conclusions and limits extraneous (outside) effects = higher quality research.

STUDY CONTROLS - CONTINUED AGAIN

• Manipulation of the Independent variable (IV):
  • In quantitative research, only one group in the study receives the intervention (IV) to explore its effect or impact.
    • Group that receives the intervention – Experimental group.
    • Group that does not receive the intervention – Control group.
  • looking to research the differences

STUDY CONTROLS - RANDOMIZATION

• Randomization:
  • Sampling process where each participant has an equal chance of being in the experimental or control group.
  • Example: 100 participants in the study are randomly assigned.
    • 50 participants to the control group and 50 participants to the experimental group.
  • Can be done by computer programs.
  • Also, randomize items on a questionnaire – especially helpful when the instrument (questionnaire) is used more than once.
  • Example – pre-test post-test research design to evaluate learning.

STUDY DESIGNS

• When examining (critiquing) a study, controls must be reviewed to ensure the accuracy of the results/findings.
• In Experimental design:
  • Testing variables, especially cause and effect relationships, it is essential to have as strict controls as possible to limit extraneous variables.

THREATS TO RESEARCH DESIGN

• Research studies must be creditable and dependable in order to inform practice.
• When critiquing research studies the consumers must examine:
  • Internal validity.
  • External validity.

INTERNAL VALIDITY

• Internal validity:
  • Is the degree to which the experimental intervention results in observable effects.
  • The researcher must rule out other factors or threats that impact the relationship between the variables.
  • These include:
    • History threats.
    • Maturation effects.
    • Testing effect.
    • Instrumental effect.
    • Selection bias.

INTERNAL VALIDITY - THREATS

• History threats:
  • Events that are not related to the planned study but occur during the time of the study and could influence the responses of the sample (participants) to the treatment.
• Maturation effect:
  • Unplanned and unrecognized changes experienced during the study, i.e., participants aging, getting stronger and or wiser.
• Testing effect:
  • Taking the same questionnaire (test/instrument) can impact scores.
  • Example: Using the same test for the pre- and post-evaluation can impact knowledge, and or skill proficiency.

INTERNAL VALIDITY - CONTINUED

• Instrumentation threat:
  • Changes in the way/manner the IV data is gathered that could account for changes in obtained measurements.
  • Example: Studying BP – ensure the use of the same type of cuffs or that instruments are calibrated at the beginning of the study in the same manner.
  • Example: More than one person observing a data collection or procedure will need to agree on how the evaluation will occur, i.e., a detailed checklist.

INTERNAL VALIDITY - CONTINUED AGAIN

• Mortality (Attrition):
  • Subjects drop out of the study before completion, which creates a threat to the internal validity.
    • The study starts with 100 participants, and 10 drop out – still lots of participants to allow for statistical analysis of the study.
    • The study starts with 100 participants, and 80 drop out – difficult to determine if the analysis is accurate.
• Selection Bias:
  • Addresses the participants who are chosen to take part in the study.
  • Example: if studying smoking cessation strategies and only the participants who are highly motivated to stop smoking are recruited, then the study results could be biased.

EXTERNAL VALIDITY

• External validity:
  • Examines the ability of the study to be generalized to additional populations and environmental conditions.
  • Example: studying smoking cessation in adults ages 30 – 40.
    • Could these results be applied to adolescents?
  • Factors that influence the external validity include:
    • Selection effects.
    • Reactive effects.
    • Measurement effects.

EXTERNAL VALIDITY - THREATS

• Selection Effects:
  • Generalizability of the results to other populations.
  • Examine how the participants are chosen to take part in the study and how subjects are grouped together.
  • Small populations or specific sampling methods to recruit participants could impact results and generalizability.
• Reactive Effects (Hawthorne effect):
  • Examines the participant’s response to being studied.
  • Participants change their responses simply because they are part of the study, not because of the research intervention.
  • Studying efficiency in a manufacturing company, participants aware of the study – both methods had the same high level of efficacy.

EXTERNAL VALIDITY - CONTINUED

• Measurement effects:
  • The administration of a pretest in a study impacts the generalizability of the results/findings.
  • Example – studying attitudes toward caring for infectious patients.
    • Makes the participants aware of their attitudes, which may then impact their receptiveness to the intervention.
    • Therefore, it could directly impact the result of the study.

SUMMARY

• When examining research and its ability to impact your practice, you must critique the research.
• Examine the study design – is it appropriate for the research question or hypothesis?
• Look at which controls were put in place to ensure uniformity and fairness within the study.
• Does the researcher work to minimize internal and external validity threats?
• Did the researcher acknowledge the internal and external validity threats?