Week 1: Selecting the Appropriate Research Design and Addressing Ethical Considerations in Research

1. General ethical principles in research

2. Ethical considerations for research with vulnerable populations  

3. Selecting the right research design/approach for your research question

Learning Objectives

·       Summarise and apply the four key values and principles of ethical research conduct outlined by the NHMRC National Statement. 

·       Summarise and apply recommendations for conducting ethical research with vulnerable populations.

·       Describe the purpose, key characteristics, and common variations of experimental and non-experimental quantitative research designs.

·       Describe the purpose, key characteristics, and common variations of qualitative and mixed methods designs.

·       Describe the purpose and key characteristics of systematic approaches to summarising past research (SLRs, meta-analyses).

·       Evaluate and select an appropriate research approach and design for a research question.

Ethical Guidelines for the Responsible Conduct of Research

• Summarise and apply the four key values and principles of ethical research conduct outlined by the NHMRC National Statement. 

Merit and Integrity

o   Merit: quality of study

o   Integrity: intentions and conduct of researchers

-              Must have potential to contribute to knowledge and understanding, social welfare and/or individual wellbeing

-              Must be conducted by ppl with appropriate qualifications and expertise, who will conduct and report research honestly and in accordance to ethical guidelines, and will disseminate results for broader community benefit

-              If research q doesn’t have merit = unethical to burden ppts.

-              Considered when developing study design and method

Justice

-              Incorporates distributive and procedural justice

-              Ppts must be appropriate for topic

-              Method of recruitment must be fair, no unfair burden of a particular social group

-              Equal distribution of benefits in participations, ppts should not be coerced or exploited.

-              Equal and fair distribution of benefits of research to broader community

-              Considered when determining an appropriate sample

Beneficence

-              Minimising harm, balancing risks of harm against potential benefits

-              Must justify any risks or harm experienced by ppts

-              Risks must be mitigated to an extent which possible

-              Ppts must be informed of those risks before agreeing to participate

-              If risks > benefits, research must be paused to allow for modification of procedures or discontinued

-              Considered in all aspects of research

-              Will also inform choice of experimental tasks and measures, storing procedures and processing data to protect ppt anonymity and privacy, informed consent and debriefing

Respect

-              Recognition of basic, intrinsic value of human beings

-              Autonomy, privacy, confidentiality, and cultural sensitives of ppts and their communities must be protected

-              Ppts should be empowered to make informed decisions about participating in research

-              Important in selection of experimental tasks and measures, which must be culturally appropriate for ppts, informed consent, and consent withdrawal, as well as procedures to protect ppt anonymity and privacy

Additional Ethical Considerations when researching vulnerable populations

• Summarise and apply recommendations for conducting ethical research with vulnerable populations.

-              Vulnerable populations: refers to groups with diminished ability to safeguard their own interests in a research setting. They may experience SE disadvantage, stigma or discrimination.

o   Factors include age, cognitive development/impairment, power imbalances, dependence on medical care, cultural or linguistic barriers, social or economic disadvantages, a history of social oppression, exploitation or disadvantage, situational vulnerabilities

Guidelines for research with ATSI peoples or communities

-              Improve the way all researchers work with Aboriginal and Torres Strait Islander people and their communities.

-              Develop and strengthen research capabilities of Aboriginal and Torres Strait Islander people and their communities.

-              Enhance the rights of Aboriginal and Torres Strait Islander Peoples as researchers, research partners, collaborators and participants in research.

Core values

-              Spirit and integrity

o   Spirit: ongoing connection and continuity between Aboriginal and Torres Strait Islander peoples—past, present, and future generations.

o   Integrity refers to the integrity of the researchers and respect for Aboriginal and Torres Strait Islander values and cultures.

o   researchers must show a commitment to upholding all the values outlined in the guidelines.

-              Cultural Continuity

o   emphasises the role of the past, present, and future in shaping collective identity (e.g., the role of historical exploitation in shaping perceptions of the present and future). This value also emphasises the critical role of personal and collective relationships in the social lives of Aboriginal and Torres Strait Islander peoples, as well as the bonds between people and the environment.

o   researchers must understand the historical experiences that may lead Aboriginal and Torres Strait Islander peoples to perceive research as exploitative. They must recognise the importance of personal and collective relationships and the role these play in the social lives of Aboriginal and Torres Strait Islander peoples and communities.

-              Equity

o   ensuring fairness and justice in research processes and outcomes. acknowledging the unique position of Aboriginal and Torres Strait Islander peoples in Australia, including their history of colonisation and dispossession

o   researchers must recognise and value Aboriginal and Torres Strait Islander knowledge, wisdom, and resources (e.g., historical, biological, and genetic resources)

-              Respect

o   refers to regard for the welfare, rights, knowledge, skills, beliefs, perceptions, customs, and cultural heritage (both individual and collective) of the people involved in research.

o   researchers must have an understanding of the values, norms, needs, interests, and aspirations of the communities and individuals involved in the research, as well as self-awareness of their own attitudes, beliefs, and behaviours.

-              Reciprocity

o   acknowledging and valuing the contributions and knowledge of Aboriginal and Torres Strait Islander communities, and building relationships based on mutual respect, trust, and shared benefits

o   researchers are expected to engage in equitable and respectful ways with Aboriginal and Torres Strait Islander peoples, including them in the research process.

-              Responsibility

o   recognition of the social and cultural responsibilities that are valued by Aboriginal and Torres Strait Islander peoples and communities, and ensuring that the research does not impact their ability to engage in these responsibilities.

o   researchers must ensure the research does not inflict harm on Aboriginal and Torres Strait Islander individuals or communities, or on the things they value.

Frameworks for Working with CALD populations

-              Culturally and linguistically diverse (CALD) populations are groups whose culture and/or language differ from that of the dominant population in a specific country.

o   populations may include immigrants, refugees, and ethnic minorities, each with their unique cultural practices, languages, beliefs, and experiences.

-              Woodland et al., (2021): framework for culturally competent research

a)        Assemble a culturally competent team

a.        having a team with the capacity to identify barriers to participation related to language and cultural differences, as well as gender, religious, and sociopolitical issues that the research must be sensitive to

b.        involves including people in the research team who understand the language spoken by the participants.

b)        Address community need

a.        Conduct research that aims to address community needs, whether they are felt needs (identified by the community), normative needs (identified by research or professional opinion), or expressed needs (indicated by the population's use, or lack of use, of public services).

c)        Address inequities

a.        Conduct research aimed at addressing inequities (e.g., poorer social, economic, or health outcomes in the target population).

d)        Address power imbalances

a.        Address differences in power between researchers and the participants and communities impacted by the research. This can include ensuring safeguards are in place to protect participants' right to informed consent

b.        also means ensuring that the selection of research questions, methods, and dissemination of results does not reinforce negative stereotypes or contribute to stigma and discrimination.

Frameworks for working with LGBTQIA+ populations

-              Individuals who identify as lesbian, gay, bisexual, transgender, queer/questioning, intersex, asexual, or other gender and sexual diversities may face vulnerabilities due to potential stigmatization, discrimination, and historical marginalization.

-              Henrickson et al., (2021) recommends 12 principles of research with sexually and gender-diverse populations

1.        Respect the dignity of all research ppts

2.        Engage with the taxonomy and language of participants

3.        Examine assumptions about who is and is not in the sample population

4.        Assume that binarized cisgender heteronormativity will have an impact on the lived experiences of gender and sexually diverse research ppts

5.        Recognise intersectionality and its impact

6.        Acknowledge multiple epistemologies

7.        Appreciate that information from gender and sexually diverse persons and communities acts indigenously

8.        Avoid problemising or pathologising the lived experiences of gender and sexually diverse research ppts

9.        Interrogate researcher (or ethics panel member) assumptions and experiences

10.  If a participant is (legally) a young person or other dependent person, prioritise informed and voluntary consent of the research participant over the need for the consent of a guardian

11.   Ensure adequate compensation for the time participants commit to the research project

12.  Generate theory from the lives of research ppts.

Selecting the right research design

·       Describe the purpose, key characteristics, and common variations of experimental and non-experimental quantitative research designs.

Qualitative research: Experimental Designs

-              Quantitative research is research that involves data that can be represented by numbers and analysed statistically.

Purpose and key characteristics of experiments

-              Used to test cause-effect relationships

o   Manipulate variables

-              Independent and dependent variable

-              Experimental manipulation: introducing or varying conditions participants experience

-              Experimental control: minimising the effects of extraneous variables on the dependent variable(s)

Between-groups experimental designs

-              between-groups (also known as independent groups) experimental design involves dividing participants into separate groups, with each group being subjected to different conditions or treatments corresponding to different levels of the independent variable (or multiple independent variables)

-              Factorial experiments

o   a researcher tests the effect of more than one independent variable at a time, as well as how those independent variables may interact to influence the dependent variable.

Repeated measures (pre-post) experimental designs

-              Repeated measures (also known as pre-post) studies involve measuring a single group of participants at multiple time points to assess whether their behaviour changes over time.

-              Simple two condition (A-B) design

o   participants are initially observed at one time point (A) and then again at another time point (B) to observe the difference.

o   conditions at each time point are systematically manipulated by the researcher

-              A-B-A designs

o   extends the A-B design by adding another phase after the experimental manipulation that reverts back to baseline conditions

o   The sequence is baseline (A), manipulation (B), and return to baseline (A)

-              Other designs with more than two time points

o   expose participants to multiple different levels of an independent variable to observe whether behaviour changes in response to variations of the experimental manipulation.

-              Simultaneous repeated measures

o   may involve alternating back and forth between conditions.

Mixed experimental designs

-              combine elements of both between-groups and repeated-measures experimental designs

-              includes multiple groups of participants who are measured at multiple time points, helping to reduce the confounding effects of pre-existing differences and time-related effects.

-              addresses the limitations of both between-groups and repeated-measures designs

o   In between-groups designs, differences in the dependent variable might arise from pre-existing group differences rather than the experimental manipulation itself.

o   Repeated-measures designs, on the other hand, may be susceptible to time-related effects, where changes in the dependent variable could be attributed to the passage of time rather than the manipulation

Experimental control: strategies for controlling extraneous variables

-              Participant characteristics

o   Random assignment to groups: Used in between-groups studies, where all participants have an equal chance of being assigned to each group. This typically results in a random and even distribution of participant characteristics between groups.

o   Using inclusion criteria: Limiting the sample to participants with specific characteristics. For example, only people from one particular cultural background.

o   Using a repeated measures design: Since the same participants are used in all conditions, individual differences that could affect the outcome are minimised.

-              Expectancy effects

o   expectations of both researchers and participants can influence participants' behaviour. This is especially true when participants are aware of the hypotheses and the group/condition they are in (experimenter-expectancy effects and placebo effects)

o   Single-blind procedure: Participants are unaware of which experimental group they are in (or, for repeated measures studies, which condition they are experiencing), and they are unaware of the hypotheses of the study.

o   Double-blind procedures: Both participants and researchers are unaware of which experimental group participants are in (or, for repeated measures studies, which condition they are experiencing), and neither group knows the hypotheses of the study.

o   No-treatment control conditions: a control condition involves an activity or alternative stimulus to make procedures as similar as possible between groups (e.g., ensuring all groups complete the study in the same amount of time, exert the same effort, etc.)

o   Standardisation of procedures: Ensuring that all researchers follow a set protocol specifying how to interact with participants.

-              Carryover effects

o   In repeated-measures designs, changes across time points may occur due to the fact that participants are measured multiple times, rather than due to the experimental manipulations themselves.

o   Practice effects, fatigue effects, context or priming effects

o   Counterbalancing: Counterbalancing is a procedure where participants are divided into groups, and each group completes the study in a different order

When would you use an experimental design

-              research question is focused on a cause-effect relationship, and your hypothesis predicts a cause-effect relationship.

-              independent variable(s) can be systematically manipulated by the researche

-              Participants can be randomly assigned to groups, or it may be possible to use a repeated measures design.

Quantitative non-experimental designs

Purpose and defining characteristics of non-experimental research designs

-              refers to study designs that do not involve the manipulation of an independent variable.

-              measure the extent to which a phenomenon occurs in a population (i.e., prevalence), to investigate relationships between variables, or to test differences between groups when random assignment to groups is not possible.

-              Defining characteristics

o   do not involve active manipulation of variables by the researcher.

o   used to describe the extent to which a phenomenon exists or the extent to which variables co-occur naturally in the world in the absence of manipulation.

Types of non-experimental research

-              Single variable research

o   focuses on studying one variable at a time

o   can answer questions about the average levels of variables in a population and the prevalence of a phenomenon

-              Cross-sectional correlational research

o   aims to measure relationships between variables without any active manipulation by the researcher

o   useful when you are interested in variables that cannot be manipulated or when you want to analyse continuous variables rather than group differences.

o   also be used to determine whether one or more variables can predict another.

-              Longitudinal cohort research

o   involves taking repeated measurements from a single group of participants at multiple time points.

o   researchers observe participants over time without interfering with the conditions they experience.

-              Quasi-experimental research

o   comparing groups of participants; participants are sorted into groups based on pre-existing characteristics.

o   Case-control studies: initial recruitment of a case group based on a particular characteristic, followed by the recruitment of a control group that is matched to the case group on other characteristics.

Non-experimental research and the issue of causality

-              not possible to directly test or interpret results as evidence of cause-effect relationships, even if a cause-effect relationship may seem plausible.

-              The following issues prevent conclusions about causality:

o   Lack of experimental manipulation: no direct test of the effect of changing the proposed causal factor (i.e., the independent variable) on the affected variable (i.e., the dependent variable).

o   Inability to establish temporal precedence: cannot definitively establish which variable (the cause) precedes the other (the effect).

o   Third-variable problem: often the possibility that other unaccounted variables are influencing the outcome.

-              Nonetheless, non-experimental designs are used in different contexts to address different kinds of research questions

When would you use non-experimental research?

-              interested in measuring a single variable.

-              interested in non-causal relationships between variables.

-              interested in a cause-effect relationship between variables, but it is not possible to manipulate the independent variables or randomly allocate participants to groups.

-              study is exploratory

Qualitative and mixed methods research

Purpose and key characteristics of qualitative research

-              Type of Data: quotes from participants, diary entries, field notes, digital data

-              Data Collection: one-on-one interviews, focus groups, direct observation, text analysis, and the collection and interpretation of documents and artefacts.

-              Data Analysis and Reflexivity: identifying and interpreting patterns and themes within the data.

Common approaches to qualitative research

-              Narrative inquiry

o   storytelling process and the meanings people ascribe to their life experiences. It involves analysing both the structure and content of stories to understand how individuals perceive and make sense of their experiences.

-              Phenomenology

o   understand and describe the meaning of individuals' lived experiences concerning a particular phenomenon.

o   Understand from their perspective rather than a collective group

-              Grounded theory

o   involves collecting data from participants about their experiences of a phenomenon and using this data as the basis for generating new theory.

-              Ethnography

o   explore and describe cultural phenomena as experienced and interpreted by individuals within a specific culture or social group

o   involves the researcher immersing themselves in the natural environment of the participants, observing them within the sociocultural context in which they exist, and viewing the situation from the perspective of an observer who is also involved in that context.

-              Case study

o   in-depth, comprehensive understanding of a phenomenon by examining a single case or a small number of cases, often within their real-life context.

o   holistic description of both the experience of a specific phenomenon as experienced by an individual or small group, and the context surrounding that case.

Mixed-methods research

-              combines elements of both quantitative and qualitative research.

-              typically much more time- and resource-consuming than using each method separately

-              provides important benefits

o   Enriching study results: can enrich the results by using the two forms of data to complement each other, providing a more comprehensive understanding of the phenomenon

o   Triangulation: the process of comparing results from each method to see if they are consistent

o   Theory or scale development: Mixed-methods research can also be used to develop new theory or measurement scales, and then test these.

When would you use qualitative research?

-              When your research question is focused on understanding, in-depth, the subjective experiences of a small number of participants.

-              When your research is exploratory and aims to build upon or expand the understanding of a phenomenon.

-              When your research aims to develop theory, rather than test existing theory or hypotheses.

Systematic approaches to summarising past research

Systematic literature reviews (SLRs)

-              critical analysis and synthesis of past research on a topic, aiming to identify what is currently known and uncover gaps in understanding that could be explored in future research.

-              Key characteristics of SLR

o   Development of a protocol

§  entire process of conducting an SLR is planned a priori and all procedures are documented in a protocol.

§  Includes the research question, how relevant studies will be located, what data will be extracted from each study, and how the quality of studies will be evaluated.

§  Researchers may, and often do, find that they need to modify their protocol during the process of completing the review – must justify

o   Systematic search procedures

§  develop clearly defined search strategies that specify the search terms to be used and the databases (or other sources) to be searched.

o   Screening studies against eligibility criteria

§  researchers systematically review the entire list of search results and evaluate whether each article meets the eligibility criteria for inclusion in the SLR.

§  use pre-defined inclusion/exclusion criteria that specify the population of interest, relevant research designs (e.g., they may only be interested in randomised controlled experimental studies), measured variables, etc

o   Extracting and summarising data

§  extract specific data from each study and present a summary of this data in their report.

o   Evaluation of study quality

§  evaluate and report on the quality of the studies included.

§  involves assessing whether the study design was appropriate to address the research question and determining if there is a risk of bias

o   Reporting procedures

§  Follows defined guidelines

Meta-analysis

-              While SLRs present a synthesis of past research in narrative form, meta-analyses take the statistical results from previous studies and statistically aggregate these results

When would you use an SLR or Meta-analysis?

-              SLR

o   aim is to provide a comprehensive, unbiased assessment of what is currently known and unknown about a phenomenon. 

o   evaluate the quality of evidence in an existing body of literature.

o   not possible to combine the results from past research quantitatively.

-              Meta-analysis

o   interested in calculating the average magnitude of an effect across all of the studies that have investigated it.

o   interested in quantifying heterogeneity

o   want to identify statistically if there are specific conditions under which the effect does or does not occur.

o   possible to combine the results from past research quantitatively.