Social Psychology: research methods and techniques

Psicologia Social: métodos e técnicas de pesquisa

DOSSIER: ON THE II SIPS PATH

Social Psychology: research methods and techniques

Psicologia Social: métodos e técnicas de pesquisa

Marcos Emanoel Pereira^I; José Luís Álvaro^II

^IFederal University of Bahia

^IIComplutense University of Madrid

ABSTRACT

The objective of this paper is to identify the research methods adopted by researchers in the field of Social Psychology, differentiating them by considerations derived from the four epistemic dimensions. Our starting point was a study conducted to identify the theoretical references and research methods used by educators and researchers in the field of social psychology. The results presented here refer to data, obtained in the years 2011 and 2012, relating to 545 social psychologists and professors of social psychology, of which 157 responded in Portuguese and 388 in Spanish. The average age of participants was 41.5 years (standard deviation = 11.4; minimum = 21 years; maximum = 78), being 54% female and 43% male. The participants originated from 19 countries, with Spain (158), Brazil (149), Mexico (64), and Argentina (45) the most frequent. Based on the results, we sought to classify and subsequently to estimate the frequency of use of the methods, considering them based on the distribution of the researchers from two geographic regions, Latin America and the Iberian Peninsula. Since geographical distribution did not provide a consistent criterion for differentiating between methods, we tried to understand the differences by considering ultimately the theoretical approach embraced by the researcher.

Keywords: quantitative methods; qualitative methods; Social Psychology.

RESUMO

O presente artigo tem por objetivo identificar os métodos de pesquisa adotados por pesquisadores da área da Psicologia Social, diferenciando-os a partir de considerações oriundas de quatro dimensões epistêmicas. O nosso ponto de partida foi um estudo conduzido com a finalidade de identificar os referenciais teóricos e os métodos de pesquisa utilizados por professores e pesquisadores da área da psicologia social. Os resultados aqui apresentados se referem aos dados, obtidos nos anos de 2011 e 2012, concernentes a 545 psicólogos sociais e professores de psicologia social, dos quais 157 responderam na língua portuguesa e 388 em espanhol. A média de idade dos participantes foi de 41,5 anos (desvio-padrão = 11.4; mínimo = 21 anos; máximo = 78), sendo 54% do sexo feminino e 43% do sexo masculino. Os participantes se declaram originários de 19 países, sendo Espanha (158), Brasil (149), México (64) e Argentina (45) os mais frequentes. Com base nos resultados, procuramos classificar e posteriormente fazer uma estimativa da frequência de utilização dos métodos, considerando-os a partir da distribuição dos pesquisadores de duas regiões geográficas, América Latina e Península Ibérica. Dado que a distribuição geográfica não proporcionou um critério consistente para a diferenciação entre os métodos, procuramos entender as diferenças levando em consideração fundamentalmente a abordagem teórica acolhida pelo pesquisador.

Palavras-chave: métodos quantitativos; métodos qualitativos; Psicologia Social.

Object of study, theoretical perspective, and investigative method are intrinsically linked in scientific research. Social psychology, since its beginning, has been marked by dissension about the object of study, the theoretical perspective, and consequently, the methods to be adopted for the investigation of phenomena to which it should be confined (Álvaro & Garrido, 2007). Defined as a discipline guided by concerns tied to the domain of psychological phenomena, psychological social psychology has tended to adopt an individualistic perspective and to adopt experimental methods; defined as a discipline of sociological origin, sociological social psychology has tended toward greater methodological diversity, to focus on a holistic perspective and survey methods or on ethnographic methodology, particularly participative observation (House, 1977). These differences in the preference for certain methods, far from moderating over time, have become increasingly pronounced, and although it is possible to identify a certain tendency to prescribe multiple approaches, this suggestion is purely rhetorical, or seems to survive much more as a suggestion for future development than in the implementation of studies conducted by social psychologists.

The complex, multifaceted, and heterogeneous nature of the phenomena subjected to scrutiny, as well as the impact of epistemological discussions about the scientific foundation of the social sciences, can pose problems in classifying the methods adopted by social psychologists. Methods for studying social behavior can be classified, for example, according to the procedure used during data collection. In this case, the distinction is established between the methods in which the researcher, given the commitment to controlling alternative explanations, produces the phenomenon intended for study, the observational methods, in which the commitment of the researcher is mainly to preserve ecological validity, being concerned with the study of the phenomenon in the real conditions in which it emerges, and finally, the methods of self-report, directed as much to obtain, by sampling procedures, a representativeness on the opinions and attitudes of the participants, as to interpret, by establishing the means to engender intersubjectivity, speech content, or to analyze discursive strategies.

Since the data collection strategy is just one of the constituent elements of any methodological approach, this does not seem to be a consistent classification criterion. A possible distinction, which contrasts the nomothetic and idiographic approaches (Grisez, 1978), helps to illustrate the methods by a differentiation between those focused on the formulation, through the development of explanatory models, of general and universal laws of social behavior, and those devoted to understanding the phenomena in the specific geographical, historical, and cultural context in which human behaviors are located.

Another traditional way of classifying psychosocial research methods draws on the differentiation between the use of quantitative and qualitative data analysis techniques. This distinction relies on a difference between those who assume that social psychology cannot separate itself from the scientific parameters embodied in studies conducted under the aegis of the natural sciences, and those who believe that the inclusion of human beings in the universe of language requires the adoption a methodological perspective different from that accepted in the natural sciences. When choosing the method based on quantitative analysis techniques, the researcher joins a tradition of research in which concepts such as variable, control, and error play a predominant role in identifying the necessary and non-contingent relationships between the antecedent and consequent variables. This search is governed by the identification, through statistical analysis, of the variance of a phenomenon, and of the partitioning of this total variance among primary and secondary systematic variances and error variance. Qualitative methodology, in adopting a more hermeneutic perspective, steers toward a tradition of research in which concepts that allow the identification of the reasons and the motives behind the actions to play a decisive role, and the concern for measurement error is replaced by an interpretive model in which the key is to try to unravel the meaning and the sense of these actions through the development of an intersubjective relationship capable of allowing the researcher to step into the conceptual world and develop interpretations about the meaning of manifested behavior, of acts of speech, and of discourse.

In a perspective more closely tied to the nature of the relationship between the investigator and the phenomenon, methods for studying social behavior can be distinguished between those dedicated to creating the phenomenon under appropriate conditions to study it, and ex post facto methods, aimed at studying phenomena after they occur. Laboratory and field experiments meet the first criterion, while the observational and self-report methods are appropriate for the second type of study.

One final criterion that we adopted to distinguish the methods of studying social behavior is related to the justification of the arguments, being able to establish the difference between the studies in which research is justified by empirical criteria and correspondence with the reality that one intends to study, and those justified by emancipatory arguments. The discussion, in this case, is fundamentally epistemological and derives support in differentiating between those who assume that the justification must be scientific and be grounded in a correspondence between the stated hypothetical proposition and observed events in the real world, and those researchers who assume that the fundamental criterion for the justification of scientific arguments is ethical and political, and is essentially defined from the contribution of the research to the efforts to combat false consciousness and in the possible assistance to disadvantaged groups to overcome the asymmetry to which they are subjected. This distinction is important because it marks the dividing line between the advocates of a social psychology drawing from the idea that the source of validation of psychosocial knowledge and the scientific model must be based on the analysis of social reality, and the advocates of a social psychology based on the notion that we should disregard the recourse to science as a legitimate source of knowledge validation given that empirical data do not offer legitimacy to any theory about reality.

Research methods for behavior and social actions

We will try, in this section, to define the criteria for classification, identification, and description of the principal methods, noting in advance that the list presented here, although it represents the major resources for research fairly well, is far from exhaustive (Figure 1).

Laboratory experiments. this is a perspective subject to a nomothetic approach, guided by a naturalistic scientific model that requires the researcher to produce and reproduce, under controlled conditions, the phenomenon being investigated. It seeks to determine the effect of one (unifactorial designs), or more than one (factorial designs), independent variable on one (univariate designs), or more than one (multivariate designs), dependent variable, through control of the sources of variation extraneous to the study and minimization of the sources of systematic secondary variance.

Field experiments. Like the laboratory experiment, this also takes a nomothetic approach, is governed by a naturalistic scientific model, and requires a direct action by the researcher in order to produce the phenomenon being investigated, although this occurs in a space distinct from the laboratory. This decision indicates a commitment to a more ecological concern than that found in the lab environment.

Computer simulations. Computer simulations involve two classes of studies, those designed to simulate, through the use of specialized software, the possible courses of action of a social phenomenon, and those that seek to reproduce, in a computing environment, classic social psychology experiments. This is a nomothetic approach, naturalistic, and involving the development and testing of researcher implemented models in suitable computing environments. Among the resources most commonly identified for the development of computerized social systems are applications and solutions such as Netlogo, Mimose, and Swarm.

Correlational Studies. This is a nomothetic approach, naturalistic, and ex post facto intended to assess the direction and intensity of association between two or more variables in order to summarize and describe opinions, attitudes, and values, (for example survey studies), to test associative relationships between theoretical constructs, or to identify the degree of fit between models and empirical data. It depends fundamentally on the reliability and validity of questionnaires, scales, and inventories.

Observational studies. This involves identifying, through a system of records, standards of conduct ostensibly likely to be subjected to scrutiny. Even if, in self-report studies and in experimental research, different observation techniques may be used, this method is equipped with an autonomous rule, conventionally termed "observational design" (Bakeman, 2000). What is notable in this particular case are the distinctions between participative and non-participative observation, and the different modes of systematic observation. A systematic observation depends basically on three defining elements: 1) defining the categories of behaviors that can be recorded; 2) clearly defining the records system (time units or event types, for example); and 3) observer training in the use of a coding system that allows some level of agreement between observers (Coolican, 2004). In a more idiographic perspective, participative observation consists of observing a phenomenon while being situated within it (from inside), adopting the viewpoint of those who are being studied (Carrera & Fernádez Dols, 1992).

In-depth interviews. Are intended to describe and analyze reality from the interviewee's own individual experiences, interpreting them from theoretical assumptions previously defined by the researcher. Interviews can be classified as directed, which have an exploratory character, and in which the interviewer introduces a topic for the interviewee to elaborate with personal reflections, and semi-structured interviews, in which the researcher follows a previously prepared script and the interviewee responds in line with the script outlined by the researcher. This is an idiographic method with a qualitative and ex post facto basis (Valles, 2009).

Discussion groups. Are intended to assess the interaction of people (from five to nine persons) with common interests and diverging views, in order to identify similarities and differences in beliefs, representations, and opinions about a given topic. The goal is to understand, through the analysis of the participants' interaction dynamics, the broader symbolic universe of social structures. This is a qualitative and idiographic approach, designed to bring out, through group techniques, opinions and viewpoints previously assumed by the participants (Alonso, 1996; Ibáçez, 1976).

Participative research. A type of research that recognizes the decisive role played by the activity of the participant, whether in the sense of directing and identifying the purpose of actions, or as a decisive element when identifying alternative actions. It is a qualitative, idiographic approach, in which specific actions are planned and carried out (Rodríguez Gabarrón, & Hernández Landa, 1994).

Action research. Practical interventions, usually conducted in natural environments, designed to produce some type of change in an organization or institution. This is an approach in which the phenomenon of interest is produced by the researcher. It has a qualitative basis and an idiographic orientation.

Life history: This is a method that seeks to gather, in a narrative perspective, information about a person, especially about one's experiences throughout life. The narrative, which takes the different cycles of life as a common thread, may be accompanied by documents and induced situations in which reminders and memories can be more easily accessed, and this action to recall/record the most important events of life on a documental basis is used by the researcher to develop interpretations of the transformations in the meaning that person attributes to his/her own life throughout its course (Mallimaci & Béliveau, 2006; Pujadas, 2002; Sarabia, 1985). It is a qualitative approach, of idiographic orientation, and ex post facto.

Discourse analysis: Upon considering that speech can not be interpreted as the result of internal psychological processes, nor as the representation of external reality, discourse analysis seeks to assess the ways in which language is used in the context of social relations, and how these discursive structures can construct, deconstruct, and reconstruct social reality. This approach is qualitative, idiographic, and grounded in the perspective of the social construction of reality. (Potter, 2003; Potter & Wetherell, 1987)

Classification of methods

To categorize the methods we conducted a hierarchical cluster analysis in order to classify the methods adopted by the participants. We conducted a cluster analysis using the proximity matrix of squared Euclidean distances based on the responses to the eleven methods submitted to evaluation. The dendrogram presented in Figure 2 allows us to identify a solution with two clusters, one which includes participative research, action research, life history, discussion groups, in-depth interview, discourse analysis, and observational studies, while the second cluster includes laboratory experiments, computer simulations, field experiments, and correlational studies.

Frequency of use of the methods

In this section we discuss the frequency of use of the above methods. Initially we thought of taking the language, Spanish or Portuguese, as a criterion for comparison, considering that they are official in the two geographical areas to be analyzed. Since we could not identify statistically significant differences in the frequencies of use, between the participants who responded to the instrument in these two languages, for any of the methods subjected to scrutiny, we preferred to present the results without this differentiation.

The graph shown in Figure 3 represent the confidence interval of the use of each of the research methods, set between the values 1, representing the alternative "never", and 4, indicating the response alternative "always", with the gray line representing the median (2.81).

Adopting the median value as criterion, we conducted a t-test for a single sample, which allowed us to distinguish three groups: the methods used with high frequency (in-depth interviews): t(242) = 9.07; p < 0.001; discussion groups: t(241) = 7.03; p < 0.001; discourse analysis: t(242) = 5.44; p < 0.001; observational studies: t(240) = 3.78; p < 0.001; participative research: t(240) = 2.70; p < 0.05), those with average frequency (life history: t(240) = 0.49; p = 0.627; action research: t(243) = 0.11; p = 0.911; correlational studies: t(236) = 1.02; p = 0.308), and those with low frequency of use (field experiments): t(228) = 6.77; p < 0.001; laboratory experiments: t(229) = 11.26; p < 0.001); computer simulations: t(234) = 13.24; p < 0.001).

High frequency. The methods used most frequently by participants were the in-depth interview (3.57), followed by discussion groups (3.35), discourse analysis (3.28), and by observational studies (3.12). Except in the case of discourse analysis, whose rationale is articulated in criteria more emancipatory than empirical, there is clearly a preference for idiographic methods of a qualitative basis and in a construction of the argument based on empirical criteria.

Average frequency. Two methods, life history (2.85) and action research (2.82) have an average frequency of use, both being linked to a qualitative and idiographic methodological perspective, while a third, correlational studies (2.73), has a naturalistic, nomothetic, and ex post facto perspective.

Low frequency. The results leave no doubt that the methods associated with a nomothetic approach, with a naturalistic scientific model, and with a perspective of construction, in an artificial or natural environment, of the phenomenon to be studied, are not of focal interest among most researchers from the Iberian Peninsula and Latin America. Field experiments (2.36), laboratory experiments (2.18), and computer simulations (2.13) are clearly nomothetic methods, linked to a quantitative model, to the perspective of creating the phenomenon to be investigated under specific conditions, and governed by an empirical justifying perspective.

Correlations between the research methods used by Portuguese and Spanish-speaking participants

Since the question presented in the survey study refers explicitly to the frequency of use of the research methods, the analyses of correlation coefficients clearly indicate, as one compares the values shown in the upper and lower diagonals of Table 1, greater use of a multi-method approach by Spanish-speaking researchers compared to Portuguese-speaking ones. This tendency is quite pronounced among the methods identified in the first cluster, all positively correlated and statistically significant, and which is seen less intensely among the Portuguese-speaking participants. As expected, the methods categorized according to the first cluster have, almost always, direct and positive associations. Some exceptions could be observed, such as the lack of association between participative research and discourse analysis among the Portuguese-language participants. Similarly, only among the Portuguese-speaking participants, no positive associations were identified between action research, life history, discussion groups, and discourse analysis. Also worthy of attention, again only among Portuguese-speaking participants, is the lack of a significant association between in-depth interviews and discourse analysis.

One also observes, as identified by cluster analysis, inversely proportional relationships between the methods assigned to the first and the second cluster. These relationships appear quite clearly between correlational studies and all the methods of the first cluster, among the Spanish-speaking participants, which did not occur among the Portuguese-speaking participants, where an inversely proportional relationship was identified between correlational studies and in-depth interviews. Among the Brazilian participants a very sharp contrast is identified between computer simulations and the action research, life history, group discussion, and in-depth interview methods.

These results together suggest that the use of multiple methods, when seen, stays within the same metatheoretical framework, and the association of methods subject to different epistemological perspectives is hardly noteworthy. Moreover, the results show, despite some differences between Portuguese- and Spanish-speaking participants, a relatively common standard of use of the methods among social psychologists from the different geographical areas included in this study.

As geography is not a distinctive criterion for differences in research method use among social psychologists, we considered another alternative interpretation, in this case, the theoretical frameworks that the participants state they adopt in conducting their research. To do this, we considered, once again from the study alluded to earlier, the intensity of the theoretical approaches adopted by the participants in conducting their research.

Relationships between research methods and theoretical frameworks

For analysis purposes, we will consider, as seen in Table 2, the following theoretical approaches: classical social psychology (classical theories in social psychology), evolutionary social psychology, group dynamics, social cognition, social identity theory, social representations, symbolic interactionism and dramaturgy school, socio-historical psychology, the Frankfurt school, community social psychology, social constructionism, approaches focused on the concept of subjectivity, rhetorical emphasis, libertarian social psychology, discursive social psychology, and historical-social approach.

Participative research. The results indicate that Spanish-speaking participants who embrace participative research tend to accept theoretical perspectives such as community social psychology, libertarian psychology, symbolic interactionism, and social-historical psychology. Notable among the Portuguese-speaking participants is the association between the use of this method and acceptance of the community social psychology perspective. A negative correlation was found between participative research and the social cognition approach, exclusively among the Portuguese-speaking participants.

Action research. The action research method was adopted predominantly by the Spanish-speaking participants who accept the theoretical perspectives of community social psychology, libertarian psychology, social-historical psychology, and symbolic interactionism. Moreover, among the Portuguese-speaking participants, in a tendency opposite the Spanish-speakers, a negative correlation was found between this method and symbolic interactionism.

Symbolic interactionism. Among the Portuguese-speaking participants a negative correlation was identified between those who embrace this perspective and action research. Among the Spanish-speaking participants, a positive correlation was identified between symbolic interactionism and all methods assigned to the first cluster, and a negative correlation with those included in the second cluster.

Life history. Among the Portuguese-speaking participants positive correlations were identified between the life history method and the theoretical approach of community psychology, while among Spanish-speaking participants, this correlation was identified, as well as additional correlations between the life history method and the approaches of symbolic interactionism, socio-historical psychology, the Frankfurt School, community social psychology, social constructionism, rhetorical focus, libertarian psychology, discursive social psychology, and the historical-social approach. Only among the Portuguese-speaking participants, was there any statistically significant correlation identified between life history and the social cognition approach.

In-depth interviews. Among the Portuguese-speaking participants a direct proportional correlation was identified between the socio-historical approach and the frequency of use of in-depth interviews. Among the Spanish-speaking participants positive correlations were identified between this method and the theoretical approaches of symbolic interactionism, the Frankfurt school, social constructionism, rhetorical focus, libertarian social psychology, discursive social psychology, and the historical-social approach, while negative correlations were identified between this method and the evolutionary social psychology and social cognition approaches.

Discussion groups. Among the Portuguese-speaking participants a positive correlation was identified between the use of discussion groups and the community psychology theoretical perspective. In the case of Spanish-speaking participants, positive and statistically significant correlations were identified between discussion groups and the theoretical approaches of group dynamics, social representations, symbolic interactionism, socio-historical psychology, the Frankfurt School, community psychology, social constructionism, approaches centered on the notion of subjectivity, rhetorical focus, libertarian social psychology, discursive social psychology, and the historical-social approach. There was no significant negative correlation relationship identified between discussion groups and any of the theoretical approaches.

Discourse analysis. Among the Portuguese-language participants a positive correlation was identified between discourse analysis and discursive social psychology. A trend similar to those mentioned earlier was identified between this method and the approaches of symbolic interactionism, the Frankfurt School, social constructionism, rhetorical focus, libertarian social psychology, discursive social psychology, and a historical-social approach. Additionally, and exclusively among Spanish-speaking participants, negative correlations were identified between discourse analysis and the evolutionary social psychology and social cognition approaches.

Observational studies. Exclusively among the Spanish participants, positive and statistically significant correlations were identified between the use of observational methods and the theoretical perspectives of group dynamics, symbolic interactionism, the Frankfurt School, community psychology, social constructionism, rhetorical focus, libertarian social psychology, discursive social psychology, and a historical-social approach. These results suggest that although we have classified the observational methods as nomothetic and quantitative, in practice the researchers who responded to this research adopt them almost exclusively in an idiographic and qualitative perspective.

Laboratory experiments. At the same time that no positive correlations were identified between the use of laboratory experiments and any of the theoretical approaches, negative correlations were identified, exclusively among the Spanish-speaking participants, between this method and the approaches of classical social psychology, social representations, symbolic interactionism, social-historical psychology, Frankfurt School, community psychology, social constructionism, approaches focused on the notion of subjectivity, rhetorical focus, libertarian psychology, discursive social psychology, and historical-social approach.

Computer simulations. Negative correlations were identified among the Spanish-speaking participants between the use of computer simulations and the approaches of classical social psychology, symbolic interactionism, social-historical psychology, Frankfurt School, community psychology, and social constructionism. No correlation of any type was identified among the Portuguese-speaking participants.

Field experiments. Among the Portuguese-speaking participants a positive correlation was identified between field experiments and the rhetorical focus. Additionally, positive correlations were identified, among the Spanish participants, between the use of field experiments and the adoption of the theoretical perspectives of classical social psychology, evolutionary social psychology, social cognition, and social identity, while negative correlations were identified between the use of field experiments and the theoretical approaches of symbolic interaction, social constructionism, libertarian psychology, and discursive social psychology.

Correlational studies. Among the Portuguese-speaking participants, positive correlations were identified between the use of correlational methods and the theoretical perspectives of evolutionary social psychology, social cognition, and the rhetorical focus. Among the Spanish-speaking participants, positive correlations were identified between correlational methods and the theoretical approaches of classical social psychology, evolutionary social psychology, social cognition, social identity, as well as negative correlations identified with symbolic interactionism, social constructionism, rhetorical focus, libertarian social psychology, discursive social psychology, and the historical-social approach.

Conclusions

What can be pointed out, first, is the existence of two large clusters regarding the use of research methods among the participants: one consisting of participative research, action research, life history, discussion groups, in-depth interviews, discourse analysis, and observational studies, and a second block consisting of field and laboratory experiments, computer simulations, and correlational studies. Concerning the frequency of use, it is clear, both among the Portuguese-speaking participants and among the Spanish-speakers, that the methods assigned to the first block are used more intensively than the nomothetic and quantitatively oriented methods characteristic of the second cluster.

With regard to the concomitant use of methods, a strong association is seen between the methods located in a single cluster, which could be a strong indicator of the adoption of a perspective of using multiple methods. This result, however, has to be considered carefully, since a good number of negative correlations were identified between the methods allocated in different clusters, which favors the interpretation that if researchers adopt multiple methods in an ever more intense manner, this is restricted solely to the methods associated with a single methodological approach and, additionally, the more often one adopts the methods of one perspective, the lesser the tendency to use methods supposedly unrelated to the perspective with which one is most accustomed to working.

The final direction of analysis considered in this study referred to relations between the methods and the theoretical frameworks. The results obtained suggest that researchers who accept theoretical perspectives such as symbolic interactionism, social-historical psychology, the Frankfurt School, community social psychology, social constructionism, approaches focusing on the notion of subjectivity, rhetorical emphasis, libertarian social psychology, discursive social psychology, and the socio-historical approach tend to favor methods allocated to the first cluster. Consistently, the methods of the second cluster are strongly associated with the theoretical approaches of evolutionary social psychology, social cognition, and social identity. These results seem to indicate that the methodological debate that marked the early development of the social sciences is ongoing and that little progress has been made toward overcoming this. It is clear that the methodological guidelines found here reflect methodological postures that are mutually exclusive, linked to diverging, and in some cases opposite, epistemological concepts. Social psychologists who prefer to use methods whose adoption implies acceptance, implicit or explicit, of a nomothetic-deductive design and of quantitative analysis techniques, are affiliated with a positivist conception of science that seeks objectivity and the separation between the researcher and the object of study. Conversely, those social psychologists who prefer idiographic methods accept an hermeneutical conception of social psychology as a social science, tend to favor inductive reasoning, emphasizing the understanding that the ultimate goal of social psychology is the recognition of the symbolic character of human action. Ultimately, it is the old debate about whether the social sciences should replicate the methods of the natural sciences (thesis of the unity of science) or, on the contrary, should be guided by the particular characteristics of their object of study. We believe that this debate is somewhat misleading, especially when it arises in terms of dichotomous and mutually exclusive models. Several examples may serve to illustrate the idea that the acceptance of a theory should not lead automatically to ascription to a particular research method. From the Frankfurt School perspective, studies were originated based on scales and on correlational analysis, for example from the studies on the Authoritarian Personality; from the perspective of libertarian psychology, Martin Baró himself proposed the creation of an institute of public opinion based on the use of surveys as the methodological presupposition for the analysis of social reality. In the perspective of symbolic interactionism, the structuralist approach of S. Stryker, in line with the Iowa School, advocates for the use of correlational methods, compared to positions such as Blumer's, a defender of qualitative methods. Likewise, scholars of social representations use various methods, included among them discourse analysis, interviews, surveys, and scales, and they conduct multivariate data analyses. These examples are sufficient to indicate that the proximity of certain theoretical assumptions should not impose an obligation to assume an exclusive relationship with the methods that are customarily used by researchers from this approach. The results of this study, however, indicate that much remains to be done to fully achieve this state, as it has become clear that methodological pluralism is only applicable when using the methods relatively close to one another.

In conclusion, it is important to note that extra care must be taken when interpreting the results, since they do not relate strictly to the actual use of methods or theoretical approaches, but to their use as is reported by participants through an instrument of self-reporting. Thus understood, we are talking less about the reality of the methods and much more about the discourse on methods. If there is any congruence between what is said and what is done, only studies that assess the congruence between what is stated and reality as it presents itself is what can clarify this issue.

Received: 05 December 2012

Revised: 27 December 2012

Accepted: 15 January 2013

Marcos Emanoel Pereira, associate professor at Federal University of Bahia and CNPq Research Scholar. Works in topics such as stereotypes, beliefs, perceptions and social prejudices. Currently is head of the Laboratory for the Study of Social and Psychological Processes (LEPPS) Psychology, at Federal University of Bahia. Email: [e-mail]

José Luis Alvaro, professor of Social Psychology at the Complutense University in Madrid. Author and co-author of numerous books, book chapters and scientific articles in the field of Social Psychology. Fellow of Churchill College at the University of Cambridge where he has been visiting schoolar and affiliated lecturer.

• Álvaro, J. L., & Garrido, A. (2007). Psicologia social. Perspectivas psicológicas e sociológicas São Paulo: Mcgraw-Hill.
• Alonso, L. L. (1996). El grupo de discusión en su práctica: Memoria social, intertextualidade, y acción comunicativa. Revista Internacional de Sociología, 13, 5-36.
• Bakeman, R. (2000). Behavioral observation and coding. In H. T. Reis & C. H. Judd (Eds.), Handbook of research methods in social and personality psychology (pp. 138-159). Cambridge: Cambridge University Press.
• Carrera, P., & Fernández Dols, J. M. (1992). La observación. Cuestiones previas. In M. Clemente (Ed.), Psicologia social. Modelos y técnicas de investigación (pp. 208-245). Madrid: Eudema.
• Ibáçez, M. (1976). Más allá de la sociologia. El grupo de discusión: técnica y crítica Madrid: Siglo XXI.
• Coolican, H. (2004). Research methods and statistics in psychology (5th edition). London: Hodder.
• Grisez, J. (1978). Métodos da psicologia social (Á. Weissenberg, Trans.). Rio de Janeiro: Zahar Editores.
• House, J. (1977). Three faces of social psychology. Sociometry, 40, 161-177
• Mallimaci, F., & Béliveau, G. (2006). Historias de vida y método biográfico. In F. Mallimaci & G. Béliveau (Eds.), Estrategia de investigación cualitativa (pp. 175-212) Barcelona: Gedisa.
• Potter, J. (2003). Discursive psychology: Between method and paradigm. Discourse & Society. 14, 783-794.
• Potter, J., & Wetherell, M. (1987). Discourse and social psychology. Beyond attitudes and behaviour London: Sage.
• Pujadas, J. J. (2002). El método biográfico: El uso de las historias de vida en ciencias sociales Madrid: Centro de Investigaciones Sociológicas.
• Rodrigues Gabarrón, L., & Hernández Landa, L (1994). Investigación participativa. Cadernos Metodológicos, 10 Madrid: Centro de Investigaciones Sociológicas.
• Sarabia, B. (1985). Las historias de vida. Revista Espaçola de investigaciones sociológicas, 29, 165-186.
• Valles, M. S. (2009). Entrevistas cualitativas Madrid: Centro de Investigaciones Sociológicas

Publication Dates

• Publication in this collection
  04 July 2019
• Date of issue
  Mar 2013

History

• Received
  05 Dec 2012
• Accepted
  15 Jan 2013
• Reviewed
  27 Dec 2012