Quanti - Research Design Flashcards
2What is a research design?
The researcher’s overall plan for
- Answering the research questions
- Testing the research hypothesis
- Used to show how all of the major parts of the research project – the samples, measures, treatments or intervention – work together to try to address the research questions
Key Research Design Features
- Intervention
- will there be an intervention - Comparison
- what types of comparisons will be made? - Control over confounding variables
- how will confounding variables be controlled? - Time frames
- how often will data be collected? - Relative timing
- when will information on IV and DV be collected?
- looking backward or forward? - Location
- where will the study take place?
Types of quantitative research designs
- Control over independent variable?
* Experimental
* Quasi-experimental
* Non-experimental - Data Collection - How often
* Cross-sectional
* Longitudinal - Data Collection - when
* Retrospective
* Prospective
What is the difference between a cross-sectional and longitudinal research design?
- Cross-sectional design:
- Data are collected at one point in time
- Longitudinal design:
- Data are collected at two or more points over an extended period
Is this a longitudinal or cross-sectional design?
Assess patients’ QoL at 1 week after their heart attack (using WHOQoL scale)
Cross-sectional
Is this a longitudinal or cross-sectional design?
Assess patients’ QoL at 1 week after their heart attack and again at 3 months and at 12 months after their heart attack (using WHOQoL scale)
Longitudinal
What are the strengths and weaknesses of cross-sectional designs?
- Are appropriate for describing phenomena at a fixed point;
- Are economical and easy to manage;
- But pose problems for inferring changes over time!
What are the strengths and weaknesses of longitudinal designs?
- Are useful for studying changes over time and for establishing sequencing of a phenomena;
- Loss of participants (attrition) over time; !
- Risk of faulty inferences.!
What is a retrospective research design?
Examine what has happened in the past.
- Data begins with outcome (DV) and looks backward for
cause or influence(IV).
What is a prospective research design?
- Investigate the outcome (DV) in the future.
- Data begins with IV and looks forward for the effect.
Is this a retrospective or prospective research design?
Examine the causes of emergency admission in a public hospital in Singapore
Retrospective
Is this a retrospective or prospective research design?
Examine the effects of a counseling service on the extent of marital problems.
Prospective
Strengths and weaknesses of a retrospective research design
- Relatively inexpensive
- Smaller numbers required
- Quicker to complete
- Prone to recall/retrospective bias!
Strengths and weaknesses of a prospective research design
- Yields true incidence rates and relative risks
- Expensive
- Requires large numbers
- Takes a long time to
complete - Prone to attrition bias!
What types of research designs involves control over the independent variable?
True experimental (RCT)
Quasi-experimental
Non-experimental
What is a Randomised Controlled Trial (RCT)?
Most powerful approach to test a cause-and- effect
relationship between Intervention (IV) and Outcome
(DV)
What is a pretest-posttest design?
- Random assignment of subjects to experimental and control group
- Outcomes (DV) measured before the intervention
- Intervention (IV) administered to experimental group
- Outcomes (DV) measured after intervention
What are the characteristics of a true experiment?!
- Intervention
- Manipulation of independent variable - Control
- Recruitment of a control group - Randomisation
– Assignment of participants to a control or experimental condition on a random basis
What is randomisation in a true experimental research?
- process by which allocation of subjects to control or
experimental groups is done BY CHANCE - Every subject has an EQUAL CHANCE of being included in
either group - Randomly assigned groups are to be comparable with
respect to an infinite number of biological, psychological
and social traits (intrinsic factors, e.g., age) at outset of the study - Group differences on outcomes (DV) observed can
therefore be inferred as being caused by the intervention
(IV)
What are some methods of randomisation?
- Flipping a coin
- Pulling names from an envelope
- Table of random numbers
- Researcher randomizer
https://www.randomizer.org/
What are the strengths of a true experimental design (RCT)?
- Most powerful design for testing hypotheses of cause-and-effect relationships.
- “gold standard” for intervention studies
– manipulation: IV is consciously manipulated by offering to one group
– randomization: ensure the study groups are equivalent and comparable in terms of intrinsic factors (e.g., age)
– control: provide a basis for comparison as both groups are expose to same external factors
- The difference in DV between two groups is due
to IV
What are the key characteristics of a Quasi-experimental research?
- Intervention: Manipulation of IV
- ABSENCE OF RANDOMISATION OF SUBJECTS
- Control (may/may not have): Recruitment of a control group
What are the strengths of a Quasi-Experimental Research?
Practical
- It is difficult or impossible to deliver an innovative treatment randomly to some people but not to others
- People are not always willing to be randomized in clinical
trials
What are the weaknesses of a Quasi-Experimental Research?
- Weaken the cause-and-effect relationship
- Absence of randomization
implied change in DV = effect of IV + initial group difference in internal factor
- Absence of a control group
implied change in DV = effect of IV + effect of unknown
external factor
What is a non-experimental research?
when researchers do not intervene by controlling the IV
In what situations will it be impossible to control the IV in a research?
- The IV inherently cannot be manipulated
e.g. Do birth weights under 1,500g cause developmental delays in children? - Unethical to manipulate IV
e.g. “What are the effects of smoking on lung cancer
development among adults in Singapore? - Research aims to investigate a natural phenomenon
e.g. “What are the quality of life and anxiety level of
patients with heart attack?
What are the Types of Non-experimental study designs?
- Descriptive
- Descriptive correlational
- Comparative descriptive
What is a non-experimental descriptive study?
To observe, describe and document aspects of a situation (a very low level research)
e.g. Examine the quality of life and anxiety level of patients with CHD
What is a non-experimental descriptive correlational study?
(more commonly used)
To describe and investigate relationships among variables, without attempting to infer causal relationships
e.g. Examine the relationship between medication adherence and quality of life in patients with CHD.
What is a non-experimental comparative descriptive research?
- Comparing & contrasting two or more samples of study subjects on one or more variables, often at a single point of time.
- Used to compare two distinct groups on the basis of
selected attributes such as knowledge level, attitudes,
physical or psychological symptoms ect.
e.g. “A comparative study on health-related quality of life and anxiety and depression levels in MI patients and Diabetes patients”
Advantages of non-experimental research
- Practical and feasible:
Human characteristics are
inherently not subject to
manipulation (e.g ethnicity,
personality, blood type);
Manipulation of IV is unethical. - Low-effort: need only hand
out the surveys and collect
the data!
Disadvantages of non-experimental research
- Shallow: data may be
superficial and fails to produce an adequate amount of data - Unable to find cause and effect relationships!!
What is a posttest-only control group design?
Excludes the use of a pretest
What is a one group pretest-posttest design?
One group: No randomisation, no control group
DV are observed before and after the administration of the tested intervention
What is a time series design (self-study)?
Monitor the DV over an extended period of time before and after the introduction of IV
Summary quiz:
What are the research designs where there IS control over the IV?
True experimental (RCT), Quasi-experimental research
Summary quiz:
What are the research designs where there is NO control over the IV?
Non-experimental research
Summary quiz:
What are the research designs where there IS control over the IV, but NO randomisation?
Quasi experimental
Summary quiz:
What are the research designs where there IS control over the IV with randomisation and control groups?
True experimental (RCT)
Summary quiz:
What are the types of true experimental (RCT) designs?
Pretest-posttest control group
Posttest-only control group
Summary quiz:
What are the types of Quasi-experimental designs?
Non-equivalent control group pretest-posttest
One group pretest-posttest
Time series design
Summary quiz:
What are the types of non-experimental designs?
Descriptive
Descriptive correlational
Comparative descriptive
