Experimental Design

Coral Imagery

Looking for in-depth understanding?

Simple research problems can generally be solved by the researcher. More complex research problems may require expert assistance to formulate the question so that it can be tested empirically. Research Design Associates provides exactly this expertise. With grounded theory and methods, we provide real-world context for Big Data patterns.

When the numbers matter Research Design Associates have methodological and statistical expertise to produce relevant answers.

Principles of experimental design are embedded in the logic of the scientific method.  While causality is not easy to demonstrate, factual outcomes come from well-designed experiments, careful observation, and systematic analysis of data.  Trustworthiness of the outcome depends on rigorous methods.

Our scientists help clients at each step of the research design process:

Formulate a measurable hypothesis that accurately reflects your research question

Plan and implement well-designed experiments

Select or develop reliable and valid measurement tools

Analyze data with statistical tests that fit the experimental design, as well as meet the mathematical assumptions of the statistics

Integrating with AI into the contextual environment

Accurately report results

Interface with AI

RDA uses empirical research to bridge Big Data and content experts.

We integrate the new algorithmic tools with the classical research methods to extend the scientific method.

Artificial Intelligence

Case Studies

Quality Not Quantity

Big Data research needs to be guided by content experts. Physicians hypothesized that getting outpatient medications and instructions to patients prior to discharge would reduce readmissions.

Read more …

Qualitative Research

Ever wonder why people make the choices they do?

Research Design Associates uses qualitative research to gather an in-depth understanding of why humans do the things they do.

Qualitative research provides an in-depth understanding of social phenomena in its natural setting. It focuses on the "why" rather than the "what" of social phenomena and relies on direct experiences. RDA researchers use multiple methods of qualitative inquiry such as in-depth interviews and focus groups to better understand the behavior, beliefs, and opinions of people. Qualitative research helps organizations develop holistic and contextual understanding about:

  • Consumer behavior
  • Decision-making in socially oriented problems
  • Non-profits’ issues, products, policies and services

Case Studies

Qualitative Research to Understand the Needs of Mentally Ill Homeless

In an effort to determine the residential, employment and service needs of Atlanta's homeless mentally ill, Research Design Associates joined with Atlanta-area hospitals, jails, shelters, and agencies that bring health care to the homeless.

Read more …

Quantitative Research

Need to make sense out of your numbers?

Analyses of Big Data can produce correlations but not demonstrate causation.

Some Big Data correlations are just random while some can be meaningful. Only well-designed studies with good statistical methods can infer causality.

RDA uses scientific investigation of numerical properties, phenomena and relationships using statistical methods. We design our research to:

  • Test hypotheses
  • Manipulate and control variables
  • Collect empirical data
  • Mathematically evaluate results
  • Conduct data modeling and analysis
  • Critically evaluate results

Case Studies

Using Quantitative Research To Solve a Complex Problem

Exercise induced heat exhaustion is a problem for humans and horses in competition when high heat and humidity compromise evaporative cooling. RDA designed and executed field studies of aggressive cooling.

Read more …

Does Big Data Mean the End of the Scientific Method?

It is well-known that correlation does not imply causation. Big Data correlation models do not replace specific hypothesis testing that can be generalized. A Big Data discovered correlation can have a numerical basis but without meaning.

Read more …

Basic Statistical Analysis

How can I find data trends that translate into valuable decision making tools?

Using sophisticated tools, RDA analysts can provide clients with enhanced insights into their research questions.

A crucial step in any data analysis is understanding the mathematical description of the data set. RDA scientists have extensive expertise in matching statistical methodologies to requirements of the research design.

Case Studies

Univariate & Bivariate Analysis

Univariate data consists of only one variable. The analysis of univariate data is the simplest form of analysis since the information deals with only one quantity that changes.

Read more …

Multivariate Research

Multivariate data involves three or more variables.  Multivariate analysis allows for analysis of more than two variables at a time.

Read more …

Mixed Methods

Is my data too complicated to be analyzed by statistics?

RDA can help design research to make complicated questions measurable.

Mixed methods research combines elements of quantitative research and qualitative research in order to answer your research questions. Mixed methods help you gain a more complete picture than a standalone quantitative or qualitative study as it integrates the benefits of both methods.

Case Studies

Can Technology Facilitate Foreign Language Learning?

Research Design Associates joined with other scientists to investigate technological innovations such as use of the internet for the listening and grammar performance of college French students.

Read more …

How Can 260 Eyewitnesses Be Wrong?

RDA experts evaluate eyewitness accuracy in the justice system using mixed model research. Eyewitness misidentification is the single greatest cause of wrongful convictions. DNA evidence has exonerated 375 wrongly convicted persons.

Read more …

Empirical Assessment of Fairness of Police Lineups

A witness  saw a robber in the dark for a few seconds. Six months later the eyewitness identified the defendant from a 6-person photographic lineup. To test the lineup used in this case, Research Design Associates researchers presented 40 people (unfamiliar with the suspect) with a description of the suspect.

Read more …

Bias in AI

Facial recognition is at the forefront of Big Data and AI. Facial recognition is a subfield of artificial intelligence in which a software algorithm is used to identify an individual’s identity by processing a digital image of an individual’s face.

Read more …

Cross-Sectional Design

Collect insights from a target audience at a particular time interval.

Cross-sectional research helps researchers collect information in a brief period.

Cross-sectional research is implemented in various sectors such as retail, education, healthcare, small and mid-size businesses. Cross-sectional research can either be descriptive or inferential and researchers rely on this survey research method in situations where descriptive analysis of a subject is required.

Case Studies

How Can Cross-Sectional Design Estimate Incidence Rates?

RDA assisted the Emory University School of Nursing in assessing the incidence of neonatal herpes in the U.S.

Read more …

Longitudinal Design

Examine changes in the same individuals that occur over a long period of time.

These studies can extend over years or lifetimes.

Longitudinal studies can be descriptive in which researchers observe and collect data on a number of variables without trying to influence those variables. They also can involve an intervention in which a group of individuals receive the treatment and another group does not.  Differences between the groups on variables of interest are measured over time.

Case Studies

10-Year Longitudinal Study of Children's Dreams

RDA scientists were involved in the analysis of this seminal longitudinal research which demonstrated that dreaming is thinking.

Read more …

Complex Designs

Complex designs allow the study of variables alone or in combination.

RDA uses statistics such as linear mixed models, path analysis, and structural equation modeling to analyze these more complex designs.

A researcher may want to study the effects of a new cognitive therapy and a drug on depression.  It may be that cognitive therapy alone has no effect on depression but when combined with a drug it alleviates depression more effectively than the drug alone.

Some research designs involve clusters of participants, organized at more than one level such as:

  • students within classrooms within schools
  • patients within units within hospitals
  • employees within departments

Special statistical methods are needed to handle analyses with individual-level effects (such as students) as well as cluster-level effects (classroom) in the same study.

Case Studies

How Profoundly Does Neighborhood Poverty Endanger the Well-Being & Development of Children & Adolescents?

Research Design Associates scientists collaborated with colleagues at Columbia University, Emory University, and Northwestern University in a longitudinal study of African American early adolescents in Atlanta, Baltimore, New York City, and Washington, D.C.

Read more …

Meta-analysis

Synthesizing research results by using various statistical methods to retrieve, select, and combine results from previous separate but related studies

RDA is experienced in conducting meta-analytic studies in a variety of content areas.

Meta-analysis is a statistical technique that applies only to research studies that generate quantitative measurement of variables and report descriptive or inferential statistics to summarize the data.

Case Studies

Mirror, Mirror on the Wall

Research Design Associates scientists worked with Emory’s University’s Department of Theater and Dance to investigate female body image.  We looked at the impact of teaching dance in front of mirrors.

Read more …

Sample Selection

A critical part of any research is the selection of a sample

Research Design Associates can resolve the issues surrounding the selection of an appropriate sample.

The meaningfulness of research results depends on the quality of the sample studied.  Accomplishing a random and representative sample of sufficient size to accurately describe the population can be a bit tricky. Research Design Associates can do it with statistical precision and confidence.

Research Design Associates routinely selects accurate sample to address a broad variety of research questions. This allows organizations to extensively examine a subset of their population to obtain factual and actionable predictions about how the larger population will perform. Sample selection is one of the few parts of research that most organizations have difficulty performing in-house. If the sample is inappropriate its use as a basis for decisions may be faulty and the resources invested in the research could be wasted.

Case Studies

Why Is Sample Selection Valuable?

If you wanted to know how tall the average male was in the US. You could measure all 150 million guys. During the study period there would be births and deaths while you were accomplishing the measurements so you would have to account for that.

Read more …

Facial Recognition Research

Can you identify the robber?

RDA experts evaluate facial recognition accuracy using a variety of experimental research designs.

Our research psychologists have been working in the field of facial recognition for over 40 years.  They have focused their research on identification procedures used in the criminal justice system and have cultivated a deep knowledge of the specific factors that influence the accuracy of facial identification.

Case Studies

How Can 260 Eyewitnesses Be Wrong?

RDA experts evaluate eyewitness accuracy in the justice system using mixed model research. Eyewitness misidentification is the single greatest cause of wrongful convictions. DNA evidence has exonerated 375 wrongly convicted persons.

Read more …

Behavioral Psychotropic Research

Researching the therapeutic effects of consciousness-expanding medicines

Psychedelic compounds like LSD and psilocybin mushrooms have shown significant promise in treating a range of mental health conditions. Participants in clinical studies often describe progress in dealing with depression and anxiety.

Psychedelics induce the brain to change transiently in ways that appear to allow a reset to take place and permit alterations in previously “stuck” ways of feeling and thinking about things.

The promise of benefits of mind-altering compounds to treat mental health problems has generated a cascade of new research.

Siegel, A.N., Meshkat, S., Benitah, K., Lipsitz, O., Gill, H., Lui, L.M.W., et al. (2021). Registered clinical studies investigating psychedelic drugs for psychiatric disorders. Journal of Psychiatric Research, 139, 71-81.

Case Studies

Is There Magic in Magic Mushrooms?

RDA scientists recently published a survey-based study that suggests that naturalistic psychedelic use provides mental health benefits similar to those observed in clinical trials.

Read more …

Contact Us

To take advantage of all that Research Design Associates has to offer please use the contact form below or call us at 404.373.4637.