Research Design: Definition and Objectives

Research Design Definition

• A research design involves a series of advanced decisions forming the master plan for a study.
• It specifies the methods and procedures for collecting and analyzing the necessary information.

Importance of Research Design

• Good research design is the "first rule of good research"
• Planning allows projects to be conducted in less time
• Preplanning results in cost savings due to efficiencies.

Objectives of Research Design

• To gain background information
• To develop hypotheses
• To measure the state of a variable of interest
• To test hypotheses that specify relationships between variables

Quantitative Research

• Originated in positivism
• Positivists believe general principles govern the social world; researchers discover these principles through objective procedures to understand human behavior.
• Uses objective measurement in controlled settings.
• Gathers numeric data to answer questions or test hypotheses.

Qualitative Research

• Based on a distinct philosophical approach that sees social reality as unique.
• Researchers understand human behavior by focusing on the meanings that events have for the people involved.
• Focuses on what people do and how they think and feel.
• Uses varied forms of inquiry to understand social phenomena from the perspective of human participants in natural settings.

Research Designs Summary

• Quantitative Research - Experimental designs, Non-experimental designs such as surveys
• Qualitative Research - Narrative research, Phenomenology, Ethnographies, Grounded theory studies, Case study
• Mixed Methods - Sequential, Concurrent, Transformative

Quantitative Research Designs

• Non-Experimental: Descriptive, Correlational, and Survey Research Designs
• Experimental Research Design

Descriptive Research Design

• Describes the characteristics of the population or phenomenon studied.
• Describes individuals, events, or conditions as they are in nature.
• The researcher does not manipulate variables but describes the sample/variables.
• Focuses on hypothesis generation rather than testing.
• It serves as a basis for further research.

Descriptive Research Design - Sample Questions

• What is the level of motivation of students?
• What is the behavior of the students in an online class?
• How does the frequency of homework completion vary among different grade levels in a mathematics program?
• What is the perception of the buyers of a certain brand?

Correlational Research Design

• Explores and identifies relationships between two or more variables.
• Explores and quantifies relationships.
• The researcher observes and analyzes the existing relationships between variables.
• It focuses on hypothesis generation rather than testing.
• It also serves as a basis for further research.

Types of Correlational Designs

• Explanatory Design: Focuses on understanding the association between variables.
• Prediction Design: Shifts emphasis to anticipating outcomes using predictor variables.

Correlational Research Design - Sample Questions

• Do socioeconomic factors correlate with health outcomes in a specific population?
• Is there a relationship between the frequency of exercise and mental health?
• Does the amount of time spent on social media correlate with feelings of loneliness in teenagers?
• Is there a correlation between income levels and spending habits in a particular demographic?

Survey Research Design

• Procedures in which researchers administer a survey to a sample or the entire population; it describes attitudes, opinions, and behaviors.
• The researcher collects quantitative, numbered data through questionnaires.
• The researcher analyzes data statistically to describe trends and test hypotheses.
• Researchers do not experimentally manipulate conditions.

Survey Instruments

• Online / Email
• Face-to-face
• Telephone
• Text Message

Types of Survey Research Designs

• Cross-Sectional
• Longitudinal

Cross-Sectional Research

• Is a non-experimental study
• Collects data from participants at a single point in time
• Generates a snapshot of a population's characteristics, attitudes, behaviors, or conditions.

Longitudinal Research

• Is a powerful and versatile methodological approach employed in various disciplines
• Investigates changes over time
• Studying complex phenomena, identifying causal relationships, and understanding individual and group trajectories.

Experimental Research Design

• Tests an idea/practice/procedure/intervention to determine its influence on an outcome.
• Establishes possible cause and effect between independent and dependent variables.
• The researcher may observe causality or manipulate variables to test causality.
• All other variables are controlled.

Types of Experimental Designs

• Between-group designs: True-experiments, Quasi-experiments, Causal-comparative, Factorial designs.
• Within-group designs: Time-series experiments, Repeated-measures experiments, Single-subject experiments.

True Experimental Design

• Is a rigorous and strong experimental design.
• Establishes cause and effect between variables by introducing a treatment/intervention.
• The researcher randomly assigns participants to conditions.
• People in the experimental group receive the treatment, while those in the control group do not.

Quasi-Experimental Design

• Used when researchers need to use intact groups
• Establishes cause and effect between variables by introducing a treatment/intervention.
• The researcher assigns intact groups to different conditions.
• Used due to participant availability or to not disrupt learning; procedures are similar to true experiments

Causal-Comparative Design

• Also known as ex post facto or retrospective research design.
• Investigates and identifies potential cause-and-effect relationships by comparing groups with different experiences
• The researcher observes and analyzes existing conditions without manipulating the independent variable.
• It is retrospective and has a control/comparison group.
• The independent variable is not manipulated.

Causal-Comparative Research Design - Sample Questions

• Do people who experienced different types of childhood trauma exhibit variations in long-term mental health outcomes in adulthood?
• Is there a difference in job satisfaction levels among employees who have experienced varying leadership styles within the same organization?
• How does the presence or absence of a mentor impact the career success and job performance of people in the early stages of their professional development?

Factorial Design

• It is a modification of the between-group experimental design.
• It also studies the independent and simultaneous effects of treatment variables on an outcome.
• The researcher observes the effects of multiple independent variables on a dependent variable.
• It may or may not use random assignment.
• It requires more than two levels/treatment conditions.

Time-Series Design

• Used when an experimental researcher has access to only one group.
• Studies the outcome of an independent variable in a group over time.
• The researcher studies one group over time with multiple pretest and posttest measures/observations.

Repeated Measures Design

• In this design a group becomes its own control.
• It also studies the outcome of two or more different independent variables in a group.
• The researcher compares a group's performance under one experimental treatment with its performance under another.

Single Subject Design

• Focuses on a single individual to study behavior over time.
• The researcher observes the impact of an intervention on the individual.
• Data is continuously gathered to observe trends.
• Used in behavioral analysis, positive behavior support, and special/gifted education.

Qualitative Research Designs

• Narrative Research
• Phenomenology
• Ethnographies
• Grounded Theory

Narrative Research Design

• "Narrative" comes from the verb "to narrate" or "to tell (as a story) in detail"
• It also describes the lives of individuals, collects and tells stories about people's lives.
• Used when there are individuals willing to tell their stories and wanting to report their stories.

Phenomenology Research Design

• Now called Descriptive Phenomenology
• It is commonly used to describe the lived experiences of individuals.
• The researcher seeks to describe how human beings experience a certain phenomenon asking questions such as "What is this experience like?"
• It attempts to set aside biases and preconceived assumptions about human experiences.

Ethnography Research Design

• Describes, analyzes, and interprets a culture-sharing group's shared patterns of behavior, beliefs, and language, which develop over time.
• It discusses culture as related to human behavior and belief.
• Used when learning about a group provides understanding of a larger issue.
• Case studies and critical ethnographies are examples.

Grounded Theory Research Design

• To generate a theory explaining a process, action, or interaction concisely.
• This theory is a "process" theory; used to explain a process of events, activities, and interactions that occur over time.
• It is also, used when we need a broad theory or explanation of a process.

Mixed-Method Research Designs

• Combines quantitative and qualitative evidence to triangulate data.
• Types:
  • Sequential: Data collected and examined in one stage informs the next phase.
  • Concurrent: Qualitative and quantitative data are collected separately but at approximately the same time.
  • Transformative / Data Transformation: Begins with qualitative data, transforming content into numerical values for statistical analysis.

Action Research Design

• Represents the most applied and practical research design.
• Is similar to mixed methods
• Its purpose is to address a specific, practical issue.
• Obtains solutions to a problem.
• Three main characteristics of this design are:
  • It is situated in a local context and focused on a local issue.
  • The research is conducted by and for the practitioner.
  • The research results in an action or a change implemented by the practitioner in the context.