Chapter 8: Selecting a Study Design (copy)

Study Design

This encompasses the focus, the kind of approach, and the mode of inquiry you will be using.

Qualitative

Less specific, non-structured; Focuses on how to understand, explain, explore, discover and clarify situations, feelings, perceptions, attitudes, values, beliefs, and experiences of a group of people; Study design and methods overlap; Seeks agreement from respondents; Non-replicable - due to its flexibility; Explore variation and diversity in any aspect of social life.

Quantitative

Specific, well-structured; Focuses on how: to ensure accuracy in measurement and classification; Study design is distinct from methods; Does not focus on respondents alone but rather in the interpretation of the data; Replicable - for verification and assurance, has to be universal truth; Focuses on the extent of variation and diversity

Cross-sectional studies(number of contacts)

Used to find our prevalence of a: phenomenon, situation, problem, attitude or issue by taking a cross-section of the population; Overall picture of the population in the time of the study / investigation; One contact only; cannot measure change

Before and after study design(number of contacts)

Used to measure the impact / effectiveness of a programme; Two sets of cross-sectional data collection points on the same population; Two contact points only; measures difference between before and after; AKA pre-test/post-test design

Maturation effect(disadvantage of B&A)

participants may change if time lapse encompasses aging.

Reactive effect(disadvantage of B&A)

participant gets educated by using the instrument thus this alters the dependent variable.

Regression effect(disadvantage of B&A)

in between contacts, participants may feel that they have been too positive or negative in one of their pre-tests causing them to change their responses in the post-test.

Longitudinal studies(number of contacts)

Used to measure the impact / effectiveness of a program over a period; Continuing data collection points on the same population; Regular intervals of data collection over a long period of time; enhanced accuracy.

Conditioning effect(disad of longi)

the multiple contacts causes the respondents to know what's expected of them thus altering the outcomes of the responses to the questions.

Retrospective(reference period)

Investigation of a phenomenon, situation, problem, or issue from the past; Data collected is based on available data in the past or respondent recall.

Prospective(reference period)

Investigation of a phenomenon, situation, problem, attitude or outcome in the future; Establishes the outcome or what is likely to happen; Experiments are classified as prospective studies as it waits for the effect of the intervention

Retrospective-prospective (reference period)

Focuses on past trends in a phenomenon and link it into the future; Data collected retrospectively from existing records before application of intervention; Note that the same study population is being focused on; No control groups; Trend studies can become the basis before investigation

Experimental(nature of investigation)

Independent variable can be: observed, introduced, controlled, or manipulated

Non-experimental(nature of investigation)

Cause has already occurred. Cause is retrospectively linked to the outcome (effect)

Semi-experimental / Quasi-experimental (nature of investigation)

Properties of both experimental and non-experimental are observed; Ex. relationship between smoking in mothers and child birth defects.

Random Design

Applied in experimental studies; Equal and independent chance of becoming part of the treatment (study population); For multiple treatment modalities, all "treatments" have the equal and independent chance of being applied to a population

The after-only experimental design

In this design, researchers measure the outcome of interest after the experimental treatment has been administered to the participants. There is no pre-treatment measurement.

Before-and-After Experimental Design

This design involves measuring the outcome of interest both before and after the experimental treatment is administered. It allows researchers to compare the changes in the outcome over time.

Control Group Design

In this design, participants are divided into two groups: the experimental group, which receives the treatment, and the control group, which does not. The control group provides a baseline for comparison to assess the effectiveness of the treatment.

Double-Control Design

Similar to the control group design, but with two control groups. One control group may receive a placebo or no treatment, while the other control group may receive an alternative treatment. This design helps ensure that any observed effects are due to the experimental treatment and not other factors; Helps quantify the impact of extraneous variables; Implemented due to: reactive effect, maturation/regression effect, or placebo effect.

Comparative Design

In this design, researchers compare the outcomes of two or more groups that have been exposed to different treatments or conditions. It allows for comparison of the effectiveness of different interventions; May be performed as an experiment or as a non-experiment; The baseline (before) must be established to have a comparison with the after treatment

Matched Control Experimental Design

Groups are matched with which they are most comparable to; They are comparable based on: socioeconomic status, the prevalence of a certain condition or the extent of a problem in the study population; Through randomization it will be decided which one of the groups will be the control and which one is experimental

Consider this

matching increases in difficulty when carried out on more than one variable, matching on variables that are hard to measure, such as attitude or opinion, is extremely difficult, and it is difficult to choose variable for matching as previous findings may come into a play of decision making

Placebo Design

In this design, participants in the control group receive a placebo (e.g., a sugar pill) instead of the experimental treatment. This helps researchers assess the true effect of the treatment by comparing it to the placebo effect.

Cross-over Comparative Experimental Design

In this design, participants are exposed to different treatments or conditions at different times, and the outcomes are compared. Each participant acts as their own control, which helps minimize variability between participants.

Replicated Cross-sectional Design

This design involves collecting data from multiple groups or samples at one point in time. Each group represents a different population or condition, and the data are compared to identify patterns or differences.

Trend Studies

Maps change over a long period of time; Useful for forecasting future events with the use of past and present data; Can be used as a retrospective-prospective basis of reference

Cohort Studies

Based upon the existence of a common characteristics such as year of birth, graduation or marriage, within a subgroup of a population

Panel Studies

Similar to trend and cohort studies except they become 'longitudinal' in nature due to working with the same population/respondents.

Blind Study

Similar to comparable and placebo experimental designs. Applied in drug effectiveness. Study population does not know if they will receive real / fake treatment. Works to isolate placebo effect.

Case Study

"Can be an individual, a group, a community, an instance, an episode, an event, a subgroup of a population, a town, or a city"; Study population must be treated as "one entity" (case); Bounded subject / unit must either be 'very representative' or ' extremely atypical'; Very flexible and open-minded technique of data collection and analysis

Oral History

More of a method than a study design; Perceptions, experiences and accounts of an event or gathering historical knowledge as viewed by individuals

Focus groups/ Group Interview

A form of strategy where there is free exchange of thoughts between members of a group and the researcher; Used in the academe and professional field due to its low cost approach; Researchers simply needs to facilitate the flow of topics in a given session

Disadvantage of Focus Groups

if not directed properly by researcher, one's opinion may dominate the discussion reducing the possible contribution of other members

Participant Observation

Gathers information about a social interaction or phenomenon in qualitative studies; Researcher deep-dives into the group being studied to observe closely the situation, interaction, site or phenomenon; Methods used can be informal interviewing, in-depth interviewing, group discussions, previous documents, oral histories.

Holistic Research

More of a philosophy than a study design; Its the practice of looking through different lenses with regards to a situation or phenomenon to get the 'whole picture' ; to get the totality of multiple factors

Community Discussions Forums

Forums focus mainly on developing town planning options and community health programmes for a community, seeking participation of its members in resolving issues relating to traffic management, infrastructure development and determining future directions for the area, informing communities of new initiatives; A forum with low attendance is not clear representation of the whole

Reflective Journal Log

Focuses on the thoughts of the researcher, It could be anything recorded within the interest of study of the researcher; As a method of data collection on its own, it could also be used in combination with other forms of methods

Action Research(philo-guided)

Action to improve practice of resolve a problem/issue; Mostly concerns quality of service; Identify areas of concern, develop and test alternatives, and experiment with new approaches.'; Under the philosophy of 'community development'

Feminist Research

main focus is the experience and viewpoints of women. It uses research methods aimed at exploring these; It actively tries to remove or reduce the power imbalance between the researcher amd respondents; The goal of feminist research is changing the social inequality between men and women. In fact, feminist research can be classified as action research in the area of the gender inequality, using research techniques to create awareness of women's issues and concerns, and to foster action promoting equality between sexes."

Participatory and Collaborative Research Inquiry

Advocates the involvement of the research participants in the research process; Advocates increased community involvement; The researcher, apart from pursuing his/her craft becomes an organizer seeking the active participation of every community member; Can undertake both qualitative and quantitative approaches.