Research Designs (Ex-post facto and experimental) | Psychology for UPSC Optional (Notes) PDF Download

Introduction

• The following unit will provide an in-depth exploration of ex-post facto research, covering its concept, meaning, significance, and types. Research can be defined as a systematic and objective process aimed at discovering new knowledge, skills, behaviors, attitudes, values, or practices, as well as reinterpreting existing knowledge, behaviors, skills, attitudes, and values. The choice of research method may vary depending on the research goal or problem at hand.
• Ex-post facto research is conducted when the researcher aims to study the cause of an event that has already occurred as a result of that cause. For instance, if a country has experienced an economic recession and the researcher seeks to analyze the "cause" or reason behind such a recession (which is the "effect"), it would fall under ex-post facto research. This unit will cover the meaning of ex-post facto research, its characteristic features, and provide a comparison with experimental research to understand how causal relationships are inferred. Additionally, we will explore the strengths and weaknesses associated with this type of research.

Forms of Research Design

• There are two forms or designs of research based on the timing of events: prospective research design and retrospective research design. When a researcher investigates the future or potential outcomes, specifically the effects that occur after manipulating the independent variable or cause, it is referred to as prospective research. For instance, if a researcher wants to examine the effects of excessive smoking, they may ask subjects to smoke under various conditions to observe the potential development of cancer.
• On the other hand, retrospective research design involves tracing the history of events to identify the underlying causes of a particular outcome. In this design, the result has already occurred, and the researcher seeks to uncover the reasons behind its manifestation by looking backward in time. These types of results are obtained through retrospective research design. For example, if a researcher aims to investigate the causes of cancer in an individual, they may explore possible factors such as smoking habits, heredity, dietary patterns, and so on by analyzing the person's history. 

Concept and Meaning of Ex- Post Facto Research

• Ex-post facto research is a type of research where the researcher aims to identify the potential causes behind an effect that has already taken place. For instance, if a child exhibits delinquent behavior, the researcher would investigate the various events and factors that might have contributed to this behavior. Possible factors could include lack of discipline at school, family history, peer influence, or neighborhood and socialization factors. It is important to note that the researcher predicts a cause based on the observed effect, as no manipulation can be done on an effect that has already occurred due to an independent variable or cause.
• Therefore, ex-post facto research can be defined as an empirical investigation that does not involve the direct control of independent variables by the researcher, as the effects have already manifested and cannot be manipulated further. Conclusions regarding the relationship between variables are drawn without intervening or manipulating the independent or dependent variable. The term "ex-post facto" is used to describe an experiment in which the researcher examines the effect of a naturally occurring treatment after it has taken place, instead of applying a deliberate treatment. In other words, it is a study that aims to uncover pre-existing causal conditions among groups.
• However, it is crucial to acknowledge that the main challenge of ex-post facto research is the assumption that if two factors coexist, one must be the cause and the other the effect.

Characteristics of Ex Post Facto Research

Ex-post facto research, also known as "causal comparative research," possesses distinct characteristics that set it apart from other types of research. The following paragraphs outline some of these key characteristics.

• The research incorporates a control or comparison group. Given that ex-post facto research examines causes that have already resulted in effects, it is necessary for the researcher to establish a control group for comparison with the actual experimental group. This allows for the analysis of the causes behind a previously occurred event.
• The behavior, action, event, treatment, or independent variable under study cannot be manipulated or altered. Since ex-post facto research aims to predict causes based on actions that have already taken place, the researcher is unable to manipulate or change those actions or behaviors.
• The research focuses on the effects. As the researcher seeks to analyze and predict the reasons behind the occurrence of an event or phenomenon, their initial focus is on understanding the event or phenomenon that has already transpired. Only after thoroughly studying the phenomenon does the researcher attempt to identify its causes.
• The research examines the "how" and "what" aspects of an event. In order to understand the causal effects behind a phenomenon, the research primarily concentrates on the "how" and "what" factors that have contributed to the occurrence of that phenomenon.
• It explores potential effects and causes. Through ex-post facto research, the researcher strives to analyze the cause-and-effect relationship of an event, action, or behavior.

Difference between an Experimental Ex - Post Facto Research 

There is no doubt that both the experimental research and the ex-post facto research try to investigate relationships between the existing variables, the conclusions of both are logically as well as empirically valid and reliable, yet there are certain basic differences between both of them. They can be pointed out as follows:

Essentials/ Requisites for inferring causal relationships

In order to infer or conclude the cause and effect relationships, the researcher needs to take care of the following aspects: 

• Associative variation: In order to infer/ predict or conclude that a particular effect ‘y’ is a result of a cause ‘x’, there should be an association between them. Therefore, the researcher can come to a conclusion only after examining that a variation in ‘x’ yields the effect ‘y’. 
• Systematic order of events: The events or the process follows a consistent sequence or order. If variable ‘x’ causes ‘y’, then variable ‘x’ must occur before or simultaneously with ‘y’, and not after it. 
• Absence of other causes: The researcher should analyse each aspect of the causal relationship with full detail and find out the best possible cause/ reason or independent variable that has ultimately led to the effect /action or event to occur. For example, the etiology of cancer can be due to several factors. In an ex-post facto research, the researcher finds out the best possible reason that might have led to the occurrence of that disease within the person or the subject.

Steps of Ex-Post facto Research

The process of conducting ex-post facto research follows a systematic sequence, as outlined by Isaac and Michael (1971). The steps involved in this type of research are as follows:

Step 1: Problem determination

• In ex-post facto research, the researcher must identify and define the specific problem they intend to study. It is crucial to analyze and understand the problem before proceeding with the research.

Step 2: Literature review

• Before predicting causal relationships, the researcher needs to conduct a comprehensive review of relevant literature and studies related to the topic. This literature review provides valuable insights for further analysis, prediction, and conclusion of the causal relationships between the variables under investigation.

Step 3: Hypothesis formulation

• The next step in ex-post facto research is to propose potential solutions or alternatives that could have led to the observed effect. The researcher must formulate assumptions that will serve as the basis for the research hypothesis and the overall research procedure.

Step 4: Designing the approach

• Once the problem is defined and the hypothesis is formulated, the researcher proceeds to select an appropriate sample that meets the study criteria. They also determine the suitable instruments or scales for collecting the necessary information or data. After finalizing the design, the researcher analyzes the relationship between the variables under investigation.

Step 5: Research validity

• The researcher needs to assess the validity and significance of their research findings. They must carefully consider the extent to which their results are valid, significant, and contribute to the interpretation and inference-drawing process.

Step 6: Interpretation of conclusions

• Finally, the researcher analyzes, evaluates, and interprets the collected information. This step involves selecting the most plausible alternative causes that may have led to the observed effect. It is through this process that the researcher can draw meaningful conclusions.

Similarly, according to Jacobs et al. (1992: 81), the following steps are involved in conducting ex-post facto research:

• Problem statement: Clearly state the research problem.
• Group determination: Select two groups from the population that differ with respect to the variable being studied in a representative manner for the sample.
• Data collection: Employ techniques such as questionnaires, interviews, literature search, etc., to gather relevant information.
• Interpretation of findings: Based on the conclusions drawn, accept or reject the research hypothesis.
• It should be noted that although ex-post facto research is a valid method for collecting information about past events, it has limitations, and only partial control over variables is possible.

Strengths and weaknesses of ex-post facto research

No research method is without its strengths and weaknesses, and the same holds true for ex-post facto research. The strengths of ex-post facto research include its relevance in studying variables that cannot be manipulated or altered, such as certain sociological (e.g., delinquency) and educational variables (e.g., achievements). It is more advantageous than experimental research when analyzing a cause based on its effect, which is not possible in experimental research.

Ex-post facto research is also less time-consuming and more cost-effective. It allows the researcher to incorporate their personal opinions and perspectives, leading to the identification of the best possible conclusions.

However, ex-post facto research does have its weaknesses and limitations:

• Inability to manipulate independent variables: As mentioned earlier, the researcher cannot manipulate or control the independent variables being studied.
• Lack of random assignment: The researcher cannot randomly assign subjects to different groups, which may limit the control over variables and the generalizability of the findings.
• Difficulty in providing a clear explanation: The researcher may face challenges in providing a reasonable explanation for the relationship between the independent and dependent variables being studied.
• It is important to acknowledge and consider these weaknesses and limitations when conducting ex-post facto research.

Concept of post HOC Fallacy

• When attempting to establish causal relationships between variables, researchers often encounter a cognitive bias known as the post hoc fallacy. The post hoc fallacy refers to the tendency of humans to draw conclusions or make predictions based solely on the observation that two factors occur together, assuming that one is the cause and the other is the effect. For example, if we observe a correlation between delinquency and parenthood, we might mistakenly conclude that parenthood causes delinquency, when in fact it may be the child's peer group that plays a more significant role.
• It is important to recognize and avoid the post hoc fallacy when examining causal relationships in research. Simply observing a correlation does not necessarily establish a cause-and-effect relationship, as there may be other unexplored factors at play. Careful analysis and consideration of multiple variables are required to ensure accurate conclusions are drawn.