Social psychology research methods encompass the intricate strategies and techniques employed by investigators to assess and investigate various variables while formulating and examining hypotheses. The central objective of social psychology research is typically to develop and assess causal theories. The emphasis on causation lies at the heart of the field’s overarching mission: to elucidate phenomena by placing them under the purview of general causal principles and to elucidate how adverse societal circumstances can be transformed by modifying their causal precursors. Naturally, there are instances in individual research endeavors where the primary aim is simply to describe existing states of affairs or to create and refine measurement instruments, and these may not directly engage with causal inquiries. Nevertheless, it is prudent to view such studies as integral components of an overarching research endeavor devoted to the construction and refinement of causal theories.
Expanding upon this, it’s worth emphasizing that social psychology research methods encompass a wide array of tools and techniques. These include experimental designs, surveys, field studies, and observational methods, among others. Each approach has its unique strengths and limitations, making it essential for researchers to choose the most appropriate method depending on the research question and objectives.
Furthermore, the pursuit of causal explanations is a cornerstone of social psychology. Researchers seek to unravel the intricate web of factors that contribute to various social phenomena, aiming to discern the cause-and-effect relationships that underlie human behavior, attitudes, and interactions. By doing so, social psychologists strive to offer insights into the fundamental processes that govern our social world and, crucially, to propose interventions that can ameliorate societal challenges by targeting their causal roots.
In essence, social psychology is a dynamic field where rigorous research methods are harnessed to uncover the intricate causal dynamics that shape our social reality. Whether investigating the impact of social media on well-being, the origins of prejudice and discrimination, or the determinants of pro-social behavior, social psychologists are united by their commitment to understanding the causal forces that drive human social behavior and the potential for positive change through this understanding.
Social Psychology Research Methods
- Autobiographical Narratives
- Bogus Pipeline
- Content Analysis
- Control Condition
- Cross-Lagged Panel Correlation
- Demand Characteristics
- Ecological Validity
- Forced Compliance Technique
- Implicit Association Test
- Lost Letter Technique
- Mundane Realism
- Nonexperimental Designs
- Order Effects
- Path Analysis
- Placebo Effect
- Quasi-Experimental Design
- Semantic Differential
- Social Relations Model
- Sociometric Status
- Structural Equation Modeling
- Twin Studies
In assessing the effectiveness of various research approaches in advancing causal theories, Cook and Campbell (1979) introduced four critical criteria, or forms of validity, to guide researchers. These criteria are pivotal in ensuring the rigor and reliability of empirical investigations. It’s essential to recognize that while statistical conclusion validity is somewhat independent of research methods, the remaining three forms of validity are intricately intertwined with the choice of research methodologies.
- Statistical Conclusion Validity: Statistical conclusion validity hinges on the judicious application of appropriate statistical tests and the use of sufficiently robust sample sizes. This aspect of validity is less contingent on the specific research method chosen and more on the soundness of statistical analysis. It ensures that the results of a study accurately reflect the underlying data and statistical relationships.
- Internal Validity: Internal validity pertains to the degree of confidence one can have in asserting that the independent variable (referred to as ‘x’) indeed exerted a causal influence on the dependent variable (‘y’) within a particular study. It assesses the extent to which the research design effectively isolates the variable of interest and minimizes the impact of extraneous factors that could confound the results.
- Construct Validity: Construct validity assesses the extent to which concrete variables (‘x’ and ‘y’) in a study align with the abstract theoretical constructs (‘X’ and ‘Y’) they are intended to represent. In essence, it measures the confidence with which a causal relationship can be inferred between these abstract constructs. Establishing construct validity is crucial for ensuring that the empirical measures accurately capture the theoretical concepts under investigation.
- External Validity: External validity manifests in two distinct forms, contingent on the research’s intended application. In particularistic research, designed to apply directly to specific settings and populations (e.g., public opinion surveys targeting a particular state or country’s voting-age population), external validity assesses the extent to which the findings can be generalized from the sample to the broader population. In universalistic research, conducted to test general causal theories, external validity raises broader questions about the generalizability of effects across diverse groups, settings, or cultures.
The methodology employed in research is shaped by numerous factors, including the research setting, the studied population, the research design, and the data collection techniques utilized. Social psychology highly values methodological diversity, recognizing that theoretical predictions corroborated by multiple methods hold greater strength. Despite this emphasis on diversity, laboratory experiments remain the predominant and defining research method in the field. This method, employed in a substantial majority of published research, provides a controlled environment that allows researchers to manipulate variables systematically and investigate causal relationships with precision.
Laboratory Research Methods
The laboratory, as a research setting in social psychology, is distinguished by its remarkable flexibility. It serves as a canvas upon which researchers can meticulously craft and orchestrate a sequence of events tailored to their specific study objectives. The inherent advantage of conducting research in a controlled laboratory environment lies in the substantial degree of control afforded to the investigator. This control facilitates the seamless implementation of experimental designs, which necessitate precise manipulation and regulation of the experiences encountered by each participant.
In theory, the choice of research setting and the composition of the studied population are independent considerations. However, in practice, laboratory studies in social psychology often employ college students as their primary participants. This tendency stems from practical considerations, such as accessibility, convenience, and resource constraints. College students are readily available, more inclined to participate in research studies, and often represent a convenient sample for academic researchers.
Nevertheless, this predilection for utilizing well-educated and generally attentive young adults in laboratory research does introduce potential limitations, particularly concerning external validity. External validity pertains to the extent to which research findings can be generalized beyond the specific sample and settings employed in a study. Relying primarily on college students may limit the generalizability of findings, as this group may not be representative of the broader population in terms of age, education, socio-economic status, and cultural diversity.
It is important to note, however, that these limitations are not inherent to the laboratory setting itself but are more closely linked to the demographic characteristics of the participants. Social psychologists are increasingly cognizant of these limitations and are actively working to address them by diversifying their participant pools. By conducting studies with a more varied range of individuals from different age groups, backgrounds, and cultural contexts, researchers aim to enhance the external validity of their findings. This ongoing effort underscores the field’s commitment to ensuring that the insights derived from laboratory research are applicable and relevant to a broader spectrum of society.
Laboratory experiments represent the cornerstone of social psychology research, offering a controlled environment for rigorous investigations (Aronson, Ellsworth, Carlsmith, & Gonzalez, 1990). These experiments employ an experimental design in which the independent variable is systematically manipulated, resulting in the random assignment of participants to different groups. This approach ensures a high level of internal validity, as it effectively isolates the variable of interest and minimizes potential confounding factors. However, lab experimentation has faced criticism primarily related to construct validity, particularly concerning the authenticity of the experimental context.
Laboratory experiments in social psychology can be categorized into three main types (Judd, Smith, & Kidder, 1991):
- Scenario or Impact Studies: These studies often feature classic social psychology experiments. Researchers orchestrate scenarios where confederates, posing as fellow participants, deliberately make inaccurate judgments or simulate emergencies to explore hypotheses about social influence on participants’ responses. When executed effectively, scenario studies immerse participants in realistic experiences, blurring the line between experimental and real-life situations. Nevertheless, questions about construct validity may arise, as participants may question the authenticity of the scenarios, suspecting they are contrived as part of the experiment. The use of deception in such studies raises ethical and practical considerations that require careful consideration (T. D. Cook, in Judd, Smith, & Kidder, 1991).
- Judgment Studies: In these experiments, participants evaluate complex stimuli, often involving individuals or social groups, and report their judgments, evaluations, inferences, or other reactions. A common focus of judgment studies is the examination of stereotyping and prejudice, where participants assess information about an individual belonging to different social groups. The construct validity of judgment studies can vary. When participants perceive the task as meaningful and realistic, it can effectively capture their genuine reactions to individuals from different groups. However, if participants discern the research’s underlying purpose and react to it, construct validity may be compromised.
- Performance Studies: Performance studies involve tasks where participants’ performance reveals underlying thoughts or emotions being studied. These studies often do not require explicit judgments but may involve tasks such as recalling information or rapidly categorizing words as positive or negative. Social cognition research frequently employs performance studies. In these experiments, hypotheses and expected responses are often less transparent to participants. Construct validity relies more on the rigorous validation of the data collection method itself, rather than participants’ perceptions of the study’s intentions.
In summary, laboratory experiments in social psychology provide a controlled platform for exploring causal relationships and psychological phenomena. While they offer a high degree of internal validity, concerns about construct validity, stemming from the artificiality of some experimental setups and participants’ awareness of the research context, are essential considerations. Researchers continually strive to strike a balance between experimental control and ecological validity to enhance the robustness and applicability of their findings in real-world contexts.
Laboratory Nonexperimental Studies
Laboratory nonexperimental studies represent another facet of social psychology research, leveraging the controlled environment of the laboratory to observe and analyze social phenomena without employing traditional experimental designs. While experimental research is often lauded for its capacity to enhance internal validity through systematic manipulation, nonexperimental studies conducted in the laboratory serve valuable purposes as well.
In these nonexperimental laboratory studies, researchers utilize the laboratory setting to establish specific conditions that allow participants to engage in experiences or facilitate detailed observations. Here are a few examples of how the laboratory environment can be harnessed for nonexperimental research:
- Observational Studies: The laboratory can be configured to facilitate unobtrusive observations of social interactions. For instance, researchers may videotape two participants engaging in an informal get-acquainted conversation. Later, these videotapes can be meticulously coded to analyze verbal and nonverbal behaviors, shedding light on patterns of communication, social dynamics, or the emergence of specific behaviors in naturalistic settings.
- Interaction and Group Studies: Small groups of participants can be brought into the laboratory to engage in various activities, such as problem-solving tasks or group discussions. Researchers can then closely observe and code these interactions to explore topics like leadership dynamics, decision-making processes, or the emergence of social norms within informal groups.
In these scenarios, the laboratory’s inherent flexibility and control are effectively employed to create specific conditions conducive to systematic observation and analysis. While no experimental manipulations are introduced, these studies yield valuable insights into the intricacies of human behavior and social interactions in controlled yet ecologically relevant settings.
Moreover, nonexperimental laboratory studies offer a bridge between the controlled conditions of the laboratory and the complexities of real-world social interactions. They provide an avenue for researchers to explore social phenomena and generate hypotheses that can later be tested through experimental research or extended to field studies for further validation. Consequently, nonexperimental laboratory studies enrich the toolkit of social psychologists, enabling a comprehensive understanding of human behavior and social dynamics across a spectrum of research methods and settings.
Nonlaboratory Research Methods
Nonlaboratory research methods, often referred to as field research, offer a distinct approach in the realm of social psychology, where the researcher’s ability to exert control over events is notably reduced compared to laboratory settings. While experimental designs in nonlaboratory settings may be challenging to implement, these methods provide opportunities to explore a wide range of independent and dependent variables, allowing for the investigation of critical and meaningful social phenomena. However, they also come with their own set of challenges related to construct validity and the complexity of real-life contexts.
Here are some key considerations and characteristics of nonlaboratory research in social psychology:
- Construct Validity: In nonlaboratory settings, researchers can investigate important psychological variables, such as the influence of role models on behavior, the impact of bystander numbers on helping behavior, or the relationship between self-concept and psychological adjustment to a diagnosis of cancer. These studies can yield valuable insights into real-world behaviors and experiences. However, construct validity may vary depending on the specific research context, and researchers must take care to ensure that their measures accurately capture the intended theoretical constructs.
- Complexity of Real Life: Nonlaboratory research is conducted within the intricacies of real-life environments, which are inherently complex and subject to a multitude of potentially confounding variables. Researchers must grapple with the unpredictability and diversity of human behavior and social interactions in these settings.
- External Validity: While nonlaboratory research is often assumed to possess higher external validity or generalizability than laboratory research, this is not necessarily the case. Nonlaboratory settings themselves can vary significantly, from a street corner to an industrial lunchroom to a hospital emergency room, each with its unique characteristics and populations. The external validity of a study’s findings in one nonlaboratory setting does not automatically extend to other nonlaboratory contexts. Replication across diverse settings is the ultimate test of external validity, similar to laboratory findings.
In summary, nonlaboratory research methods in social psychology provide a valuable complement to laboratory research by offering insights into real-world behavior and experiences. These methods enable the exploration of crucial social phenomena but come with the challenge of navigating the complexity and unpredictability of noncontrolled settings. Researchers must be diligent in addressing construct validity and recognize that the generalizability of findings is contingent on replication across a variety of nonlaboratory contexts. Through a combination of laboratory and nonlaboratory research, social psychologists aim to develop a comprehensive understanding of human behavior and its societal implications.
Field experiments represent a powerful approach to experimental research in social psychology conducted outside the controlled confines of the laboratory. While they present logistical challenges, they offer numerous advantages and insights into real-world behaviors and social phenomena. Here are key points about field experiments:
- Realistic Context: Field experiments are designed to investigate social phenomena in natural, real-life settings. Researchers construct situations that mimic genuine circumstances, allowing for the examination of behaviors and reactions as they naturally occur. For instance, a study on bystanders offering help to a person in need may involve simulating a roadside breakdown to assess the impact of interventions, such as community responsibility billboards, on the frequency of assistance offers.
- Experimental Design: Field experiments maintain a high degree of internal validity through the use of experimental designs. This entails systematically manipulating variables of interest and randomly assigning participants (or in this case, bystanders) to different experimental conditions. Randomization ensures that groups exposed to various manipulations are equivalent, reducing the influence of confounding factors.
- Construct Validity: Field experiments benefit from construct validity because the manipulations and measurements are grounded in meaningful and realistic contexts. Researchers strive to ensure that the interventions and measures accurately capture the theoretical constructs under investigation, enhancing the relevance and authenticity of the study.
- External Validity in Particularistic Research: Field experiments conducted in settings and populations of direct interest exhibit high external validity in particularistic research. For instance, if a field experiment on the effects of different working conditions on productivity occurs within an actual industrial setting, the findings are highly applicable to that specific context and population.
- Considerations in Universalistic Research: In universalistic research, it’s essential to recognize that nonlaboratory settings vary considerably. The fact that a field experiment was conducted outside the laboratory does not automatically guarantee broad generalizability. Each unique setting and population may yield different results, emphasizing the importance of replication and contextual awareness.
In summary, field experiments in social psychology bridge the gap between controlled laboratory research and real-world social interactions. They offer the advantage of high internal validity while capturing the richness and complexity of genuine social situations. However, researchers must be cautious about the limitations of generalizability, as findings from one field experiment may not necessarily apply universally. Rigorous replication and a nuanced understanding of the specific context are crucial for drawing robust conclusions in field experiments.
Quasi-experimental designs offer a middle ground between true experimental designs and nonexperimental research methods, providing a way to investigate causal relationships with some control over variables but without the stringent demands of random assignment. Here are some key points about quasi-experiments:
- Internal Validity: Quasi-experimental designs can help mitigate certain threats to internal validity, but they do not eliminate them entirely, as true experiments do. While they lack the randomized assignment of participants to different groups, they still involve manipulation of the independent variable to some extent. This manipulation allows researchers to draw causal inferences, but the presence of confounding variables remains a concern, potentially impacting the degree of internal validity.
- Control and Implementation: Quasi-experiments impose fewer demands for strict control compared to true experiments. This flexibility can make them more practical and feasible to conduct outside the laboratory. Researchers can manipulate the independent variable in real-world settings, such as implementing an ad campaign in one city while using another comparable city as a control.
- Manipulation without Random Assignment: In quasi-experiments, manipulation of the independent variable occurs without random assignment. This means that participants are not randomly assigned to different conditions or groups, raising concerns about the equivalence of the groups and the potential influence of confounding variables. Researchers must carefully consider the potential impact of these confounds when interpreting results.
- Practical Applications: Quasi-experiments are often used in applied settings where it may be challenging or impractical to employ strict experimental control. For example, researchers may investigate the effectiveness of public health campaigns, educational interventions, or policy changes using quasi-experimental designs. These studies can provide valuable insights into the real-world impact of interventions.
- Similar Considerations to Field Experiments: Many considerations that apply to field experiments, such as construct validity and external validity, also apply to quasi-experiments. Researchers must ensure that their manipulations and measurements are meaningful and realistic. Additionally, the generalizability of findings from quasi-experiments may vary depending on the specific context and population under study.
In summary, quasi-experimental designs strike a balance between experimental control and real-world applicability. They offer a way to investigate causal relationships in settings where random assignment may not be feasible. While they can provide valuable insights into cause-and-effect relationships, researchers must be mindful of potential confounding variables and limitations in internal validity. Careful design and analysis are essential to draw meaningful conclusions from quasi-experimental research.
Survey research is a fundamental methodology employed in various social sciences, including sociology, political science, and even within the domain of social psychology. This approach is instrumental in investigating a wide array of social phenomena and understanding the attitudes, behaviors, and perceptions of individuals within a population. Here are key aspects of survey research:
- Data Collection from Representative Samples: Survey research typically involves the collection of data from a representative sample of the population of interest or, in some cases, the entire population. For instance, researchers might survey voters in a specific state or all employees of a particular company. The goal is to gather insights that can be generalized to a larger population.
- Self-Report Data Collection: Surveys rely on self-report data collection methods, where participants respond to questions or statements about their attitudes, beliefs, behaviors, or experiences. This can be accomplished through various means, including personal interviews, telephone interviews, or written self-administered questionnaires. Each data collection mode has its unique strengths and weaknesses, impacting factors like cost and data quality.
- Nonexperimental Design: Survey research is typically nonexperimental in nature, meaning that researchers do not manipulate variables as they would in experimental studies. Instead, survey questions are designed to measure existing attitudes, behaviors, or conditions. While this limits internal validity in terms of establishing causal relationships, survey research excels at exploring associations and patterns within populations.
- Construct Validity Challenges: Construct validity in survey research can be challenging due to the method of data collection, which relies on self-report. Participants may introduce biases into their responses, consciously or unconsciously, leading to potential inaccuracies. For instance, respondents may provide socially desirable answers rather than revealing their true beliefs or behaviors.
- External Validity: Surveys often demonstrate high external validity, particularly in particularistic research where the focus is on generalizing findings from the sample to a specific target population. This makes survey research a valuable tool for understanding the attitudes and behaviors of specific groups, such as voters in a state or employees within an organization.
In summary, survey research plays a vital role in social sciences by providing a means to collect data on a wide scale and explore the perspectives of individuals within a population. While it may lack the experimental control of laboratory studies, surveys offer valuable insights into social phenomena, attitudes, and behaviors. Researchers must be mindful of potential biases and limitations in construct validity but can leverage the high external validity of survey findings to inform their investigations and policymaking.
Naturalistic Observational Studies
Naturalistic observational studies are a valuable research method in social psychology that involve observing and documenting naturally occurring social behaviors in their real-world settings. These studies provide a unique perspective on human behavior by capturing actions and interactions as they naturally unfold. Here are some key characteristics of naturalistic observational studies:
- Realistic Settings and Populations: Naturalistic observational studies take place in realistic settings, often within the context of everyday life. Researchers venture into environments where people naturally engage in social behaviors, such as schools, workplaces, public spaces, or homes. This approach allows for the study of human behavior in its natural context.
- High Construct Validity: Construct validity in naturalistic observational studies is typically high. This is because measurements are taken in authentic, real-world settings and populations, providing a more genuine representation of social behaviors. Researchers have the opportunity to observe and document behaviors as they naturally occur, minimizing the influence of artificial or controlled conditions.
- Lack of Experimental Design: One key characteristic of naturalistic observational studies is the absence of experimental design. Unlike laboratory experiments, researchers do not manipulate variables or introduce interventions. Instead, they act as passive observers, documenting behaviors without interference. This lack of experimental control can limit internal validity, making it challenging to establish causal relationships between variables.
- Examples of Naturalistic Observations: An example provided in your question illustrates this approach: researchers interested in understanding the dynamics of intergroup relations in an elementary school might discreetly observe the seating patterns in the school lunchroom. By documenting where students of different racial backgrounds choose to sit, researchers gain insights into the extent of racial segregation in a natural, unaltered context.
- Rich Descriptive Data: Naturalistic observational studies yield rich descriptive data that can inform theories, generate hypotheses, and provide valuable insights into human behavior. Researchers often focus on qualitative and quantitative descriptions of behaviors, interactions, and contextual factors.
- Ethical Considerations: Ethical considerations are paramount in naturalistic observational studies. Researchers must ensure that their observations are unobtrusive and respectful of individuals’ privacy. In some cases, obtaining informed consent may be necessary, particularly when studying sensitive or private behaviors.
In summary, naturalistic observational studies offer a window into the real-world dynamics of social behavior, providing a high level of construct validity. While they lack the experimental control found in laboratory experiments, these studies contribute valuable insights to our understanding of human behavior in its natural context. Researchers must carefully balance the benefits of authentic observations with ethical considerations and the limitations in establishing causal relationships.
Analysis of Archival Data
Analysis of archival data is a research method in social psychology that involves the examination of pre-existing records or data sources to test research hypotheses and gain insights into social phenomena. This approach leverages information stored in official or unofficial archives, such as government records, newspaper articles, library circulation records, and more. Here are key characteristics and considerations related to the analysis of archival data:
- Objective and Comprehensive Coverage: Archival data provides an objective and often comprehensive view of a population of interest. Unlike self-report measures, which rely on participants’ responses, archival data can offer a complete and unfiltered representation of real-life outcomes. Researchers can draw from a wide range of sources to access information on various social phenomena.
- Examples of Archival Data Analysis: An example provided in your question illustrates this method: researchers interested in the hypothesis that heat increases aggression may examine official weather records and crime statistics. By analyzing these data sources, researchers can investigate whether there is a correlation between hot weather and an increase in homicides. Archival data analysis allows researchers to explore relationships between variables by utilizing historical or pre-existing records.
- Construct Validity Considerations: Construct validity in archival data analysis can be a concern. This issue arises when the archival measures do not directly correspond to the psychological constructs of interest. For instance, the legal definition of homicide may not precisely align with the psychological concept of aggression. Researchers must carefully assess whether the archival data accurately capture the variables under investigation.
- Nonexperimental Designs and Internal Validity: Archival data analyses typically involve nonexperimental designs, meaning that researchers do not manipulate variables or introduce interventions. As a result, the internal validity of these studies tends to be low. Establishing causal relationships between variables can be challenging in the absence of experimental control.
- Strengths and Limitations: The strengths of archival data analysis include its ability to access historical and real-world data, offering insights into social phenomena beyond self-report measures. Archival data can provide valuable information about long-term trends, patterns, and historical events. However, researchers must contend with the limitations of construct validity and the challenge of demonstrating causal relationships.
- Ethical Considerations: Ethical considerations in archival data analysis involve respecting the privacy and confidentiality of individuals whose data is included in the archives. Researchers should adhere to ethical guidelines and obtain any necessary permissions or approvals when working with archival data.
In summary, the analysis of archival data is a valuable research method in social psychology for investigating social phenomena through the examination of existing records and data sources. While it provides objective and comprehensive information, researchers must carefully assess construct validity and recognize the limitations in establishing causality due to the absence of experimental control. Archival data analysis complements other research methods and offers unique opportunities for studying historical and real-world social dynamics.
Research without Primary Data Collection
Research without primary data collection is a valuable approach in social psychology that involves the analysis and synthesis of existing studies to draw conclusions and make informed insights. This method encompasses meta-analysis and computer simulation, each serving distinct purposes within the research process.
- Meta-Analysis: Meta-analysis involves the quantitative synthesis of multiple primary studies on a specific topic. Researchers collect data from previously conducted studies and analyze it collectively to draw more precise and objective conclusions. For example, if numerous studies have explored sex differences in helping behavior, a meta-analysis can provide an overall assessment of the differences and identify factors that influence these variations. Meta-analysis enhances construct validity by examining multiple operationalizations of constructs and bolsters external validity by including various settings and participant populations.
- Computer Simulation: Computer simulation is a method for exploring the implications of a theory by creating a computer program that embodies the theory’s assumptions. The program generates predictions based on specified conditions, allowing researchers to gain insights into the theory’s outcomes in complex scenarios. Computer simulation is particularly useful when theories are too intricate for intuitive prediction. However, it is essential to note that computer simulation is not a substitute for data collection. Instead, it assists in deducing a theory’s implications, which then become research hypotheses to be tested using empirical data. If the hypotheses do not align with the data, researchers may need to modify or discard the theory.
In social psychology, laboratory experimentation remains the predominant method. However, researchers recognize the value of diverse methods and appreciate that the most robust research findings are those that can be replicated across different settings and populations, using various research techniques. This emphasis on methodological diversity ensures a more comprehensive and nuanced understanding of social phenomena.
In summary, research without primary data collection, such as meta-analysis and computer simulation, offers complementary approaches to traditional empirical studies. These methods provide valuable insights, enhance the precision of conclusions, and contribute to the robustness of research findings. The integration of multiple research methods and replication across diverse contexts strengthens the foundations of social psychology research.
- Abelson, R. P. (1995). Statistics as principled argument. Hillsdale. NJ: Erlbaum.
- Aronson. E., Ellsworth. P. C., Carlsmith, J. M., & Gonzales. M. H. (1990). Methods of research in social psychology (2nd ed.). New York: McGraw-Hill.
- Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation. Chicago: Rand McNally.
- Judd. C. M., Smith, E. R., & Kidder. L. H. (1991). Research methods in social relations (6th ed.). Fort Worth. TX: Holt, Rinehart, and Winston.
- Mook. D. G. (1980). In defense of external invalidity. American Psychologist, 38, 379-388.
- Reis, H. T., & Judd, C. M. (Eds.). (in press). Handbook of research methods in social psychology. Cambridge. UK: Cambridge University Press.