Research Design B&R Week 2

Research Aims

What is meant by research aim?

Research aim – statement that describes the purpose of the study 

• Short re-statement of the research topic or theme  

Should only be one aim and it should be communicated in one sentence- written using verbs 

What is meant by research question?

Research questions- what the study will answer after data analysis and interpretation  

• Questions must be concise to provide clear foundations of hypotheses 

What is the FINER criteria?

• F Feasible- appropriate number of participants 

• I Interesting getting the answer intrigues the investigator, peers and the community  

• N Novel confirms or extends previous findings 

• E Ethnical Consistent with the standards that will lead to ethical approval  

• R Relevant Scientific Knowledge and applied practice

What is the PICOT criteria?

• P Population What specific group are you interested in? 

• I Intervention What are you investigating  

• C Comparison What is the alternative to compare with the invention 

• Outcome What do you aim to accomplish 

• T Time what is the appropriate follow up time to assess the outcome 

What is meant by research objectives?

Objectives define specific aims and how the overall aim is completed  

• Translate overall aim into operational standards  

• Explain how outcome will be achieved 

• Explain how study is going to answer the question

What is a research hypothesis?

Clear and testable statement about possible outcomes of the study 

• Drives the research question and not driven but data itself 

Hypothesis developed from research questions and the main elements of study are summarised in a form that establishes basics for testing 

Simply predicts the relationship between a single dependent variable and independent 

What are the traditional kinds of research hypothesis?

Null hypothesis- will be no difference in resting metabolic rate between males and females  

Alternate- resting metabolic hypothesis higher among males compared to females   

Complex hypothesis- predicts the relationship between multiple dependent and independent variable   

What are the six kinds of research hypothesis?

Simple hypothesis: Predicts relationship between a single dependent variable and single independent variable

Complex hypothesis: Predicts relationship between multiple dependent and independent variables

Directional hypothesis: Identifies the direction of the relationship between variables

Non-directional hypothesis: Does not specify a direction of effect (intentionally)

Associative hypothesis: Change in one variable is associated with change in another

Causal hypothesis: Manipulating the independent variable will change the dependent variable

What is a cross-sectional study?

Assesses a phenomenon at one point in time, Examines trends and changes at a societal/national/international level

• Measures different samples/populations

What is a longitudinal study?

Assess a phenomenon at several points in time, provides information on change at the

individual level

• Measures the same sample/population over time

What is research design?

Provides framework of methods and techniques- allows research to be constructed in structed ways  

Tells how a study is being conducted- descriptive and experimental  

Observational- provides information between exposure and event or characteristics of population  

Experimental- intervention- allows for determination of the cause and effect if study designed appropriately

What are the 2 types of research design?

Retrospective Design- examines data that already exists, examines factors that predict something that has already happened  

Prospective Design- collects new data, waits for outcomes and relates this into suspected influencing factors  

What is meant by randomised control trials?

Participants are randomised into controlled, intervention groups  

Follows groups overtime to determine difference in outcomes  

What are the threats to internal validity?

Internal- whether the study design conduct and analysis the research questions without bias 

History- unintended event during the study has an influence 

Maturation- process within participants that happen due to time 

Testing- learning effect of multiple testing  

Instrumentation- changes in instrument calibration, staff performing procedures differentially 

What are the threats to external validity?

Whether the study findings can be generated to other contexts and the ‘real world’  

• Reactive or interactive effects of testing- pretesting  

• Interaction of selection bias and treatment  

• Reactive effects of experimental treatment 

• Multiple treatment interference 

What is epidemiology?

Study of the occurrence, distribution and determinants of health-related events, states and processes in specified populations and the application of this knowledge to control relevant health problems.  

Can be cross sectional, case control, cohort/longitudinal 

What is retrospective design?

Backward looking: examine data that already exist

Tries to identify factors that predict whether something will happen

• disease, sports performance

What is Prospective design?

Forward looking: collects new data, then sometimes, watch/wait

Waits for outcomes (e.g., development of disease, or sporting performance) and relates this to suspected influencing factors

What is a randomised control trial?

Forward looking (prospective)

Participants are randomised into a “control” or “intervention” group

Follows groups over time to determine a difference in outcome

How do randomised control trials show cause and effect?

The cause must precede the effect

The cause and effect must be correlated

The correlation between cause and effect cannot be explained by another variable

What are feasibility studies?

Research done before a main study to determine whether the methods are feasible and acceptable

Can this study be done?

What are Pilot studies?

Pilot studies:

A small-scale study conducted prior to a large-scale experiment to test and refine procedures

“Does initial data suggest our method/intervention could work?

What is sampling?

A “sample”: a group of participants, treatments, or situations from a larger population

• Sometimes described using italic letters (like n = 25) representing number of people

A statistic describes the sample of participants from which it was calculated

•If the sample represents a larger group then the findings can be inferred (generalised)

What are the 5 kinds of sampling?

Random sampling: randomly select participants from a larger group

Stratified random sampling: The population is divided (stratified) on a characteristic

Voluntary sampling: People volunteer via advertising

Convenience sampling: Researchers may be interested in a specific type of person

Snowball sampling: Participants recruited via convenience sampling help!

What are the two ways to assign people to groups?

Random assignment: randomly allocate participants to different study groups

Random matched assignment: considering a participant characteristic