Building on a foundation of Java programming and sociological survey methods, the proposed project will use the World Wide Web to assemble cross-cultural databases of subjective culture, and will make these databases available to multi-national and multi-disciplinary professionals via the World Wide Web. Issues covered in this proposal include: tool development, data collection of affective meanings at remote sites in and out of the U.S., storing data for world-wide retrieval, and providing on-line simulators to translate quantitative measurements of meanings into verbal interpretations in multiple languages. The proposed research relates to a suggested Knowledge-Networking topic: ’Designing tools for gathering and analyzing data, including new types of tools to collect, share, and manipulate increasingly complex data sets and structures, with underpinning innovations in computing, telecommunications, sophisticated algorithms, and hardware/software systems.’
Scientific assessment of subjective culture became a practical matter in the final quarter of the twentieth century. Psychologists Osgood, May, and Miron (1975) derived three dimensions of affective measurement from multivariate analyses of scale ratings in more than 20 cultures, and they showed that the same three dimensions of affective meanings exist cross-culturally. Sociologist Heise (1979; Smith-Lovin and Heise, 1987) mathematically modeled the human cybernetic system that operates on affective meanings in specific situations and demonstrated how the system combines culturally-given affective meanings in the construction of a variety of cultural forms such as roles, emotions, attributions, and identities. Anthropologists Romney, Weller, and Batchelder (1986) mathematically analyzed informants' knowledge about a homogeneous culture and showed that since each informant's report serves as an indicator of a single state in the culture, the human sampling requirements for a reliable assessment of that state are far less than the sample sizes required to survey multiple positions in a diverse population.
These developments imply that significant portions of cultural knowledge can be measured economically, stored in a computer, and retrieved in diverse forms to understand social relations; and the same general procedures can be applied in different societies to create knowledge databases for cross-cultural study. These are the challenges addressed in this proposal.
Semantic differentials are bipolar scales defined with contrasting adjectives at each end (Osgood, May, and Miron, 1975). For example, a scale might contrast "good, nice" with "bad, awful" and provide seven rating positions in between. The middle position is labeled "neutral" and coded zero; positions on the "good, nice" side are labeled with adverbs "slightly," "quite," "extremely" and coded with positive numbers; and positions on the "bad, awful" side are labeled with the same adverbs but coded with negative numbers. A number of considerations are involved in this technology.
1. Bipolar adjective scales are a simple, economical means for obtaining data on people's reactions to stimuli. Such scales can be administered to adults or children from any culture.
2. Three dimensions of response -Evaluation, Potency, and Activity (EPA) - account for most of the co-variation in ratings on bipolar adjective scales. The Evaluation dimension is tapped by the "good, nice" versus "bad, awful" scale just mentioned. The Potency dimension corresponds to a scale that contrasts "powerful, big" with "powerless, little." A scale for assessing the Activity dimension contrasts "fast, noisy, active" with "slow, quiet, inactive." Pan-cultural multivariate analyses have demonstrated that these EPA dimensions are clearly recognizable in multiple cultures and a variety of languages.
3. EPA measurements are appropriate when one is interested in affective meanings rather than denotative or logical meanings. Affective meanings correspond to sentiments about something - e.g., the general feelings that we have about mothers. The EPA system is notable for being a multivariate approach to affective meanings, as compared, say, to attitude measurement which deals only with the single dimension of Evaluation.
So far, EPA databases have been acquired for U.S. culture (Heise, 1978; Smith-Lovin and Heise, 1988), Canadian culture (MacKinnon, 1994), German culture (Schneider, 1997), and Japanese culture (Smith, Matsuno, and Umino, 1994). Datasets and related programs that already exist can be found on the World Wide Web at http://www.indiana.edu/~socpsy/ACT/Index.html.
Early work with semantic differentials presented stimuli and scales via paper-and-pencil questionnaires, and ratings then were coded and keyed into electronic datasets. Heise (1988) advanced the technology by developing a computer program that presented stimuli and scales in random order on a computer screen, and allowed respondents to key in more refined ratings than were permitted by nine check positions; the program automatically coded ratings and stored the data on diskettes. This program was translated to languages other than English, and used to assemble affective meaning databases in Germany and in Japan.
Recently, Heise wrote a Java applet in order to collect EPA ratings via the World Wide Web. Respondents with a computer connection to the Internet go to a WWW page that fetches the Java applet and its associated text files. The applet presents stimuli and semantic differential scales randomly, and the respondent rates the stimuli with the computer's mouse, by dragging a pointer along the scales and clicking buttons on the screen. The applet records the respondent's ratings in numerical form, along with latencies (the time elapsed from presentation of stimulus to final rating), and sends all of the data to a central computer for storage when the respondent finishes the ratings. Thus, this tool collects precise numerical data on affective meanings at remote sites, even on different continents, and collates the data at a central computer for statistical processing and archiving. The program performed successfully in a 1998 test involving EPA ratings of 25 concepts by 70 university students.
During the project period we will employ the Java applet in the U.S. to measure EPA affective-meanings of more than 2,000 concepts related to social relations. We will draw respondents from multiple regions of the U.S. by establishing data-collection stations at universities with good facilities for Internet connections and with cooperative local representatives who can recruit student volunteers and guide the subjects through a session of on-line ratings. This large project will allow us to refine our procedures for acquiring data via the World Wide Web, the project will update the only available U.S. affective-meanings database which now is two decades old, and the project will build an organizational legacy that can be used in later research projects.
Additionally we will conduct a survey of a probability sample of Indianapolis, IN, residents, in order to compare their sentiments with sentiments measured among college students from the same city. Interviewers will visit about 300 residents in their homes, and the residents -motivated by cash payments - will use portable computers to rate about 100 social concepts on semantic differentials. The 100 concepts will be chosen so as to help us assess the conditions under which the homogeneity assumption applies: i.e., when do ratings from college respondents provide adequate representations of ratings from a more general population.
Also during the project period we will employ the Java EPA applet to assemble affective-meanings databases for foreign cultures. First we will complete internationalization of the EPA applet, in particular incorporating Asian printed characters. After testing the internationalization with translations to a number of languages, our senior comparative sociologists will employ the applet to collect data via the World Wide Web from Germany and China. Additionally, with the participation of research assistants from foreign countries who are graduate students in our Departments of Sociology, we will seek to collect data from Africa, South America, Southeast Asia, the former Soviet Union, and the Middle East.
We have three goals in these data-collection projects:
1. To develop methods for conducting social surveys of remote populations via the World Wide Web;
2. To establish stations around the world with personnel trained in Internet use and procedures for aligning respondents with computers so that future surveys can be fielded easily;
3. To assemble new or more up-to-date cross-cultural databases of affective meanings.
Though the first few databases of affective meanings were published in printed form (see Heise, 1978), electronic modes of distribution are better for distributing information that is likely to be re-used in computers. In the 1980s Heise developed MS-DOS software for searching an affective-meanings database either by word or by EPA profile, and in the early 1990s Heise developed Macintosh software to allow cross-cultural comparisons and searching for words by their indigenous spelling or by their English translations. These programs and databases are offered freely on the Internet, but the platform specificity of the software limits their usefulness for researchers at other sites and in other disciplines.
In 1998 affective-meanings databases were assembled in a desktop computer and linked to the World Wide Web via a commercial database program (FileMaker Pro). This feasibility test showed that the databases can be on-line for users around the world, and researchers who are responsible for compiling a particular database can add or edit data from remote sites.
During the period of the proposed project we intend to extend the on-line availability of the affective-meaning databases in several ways.
1. In addition to the current presentation form for studying meanings within a single culture, we will provide forms for comparing the affective meaning of a concept across cultures, and for searching for concepts by EPA profile.
2. We will seek a security system to reduce the possibility of malevolent attacks on the archived data that is available on the World Wide Web. Additionally, we will develop an automatic backup system that protects against loss of data as a result of errors by researchers working in databases where they have authorized access.
3. We will internationalize the database system so that work can be conducted either in the indigenous language of an affective-meanings database or via English translations.
Affect control theory (Heise, 1979; Smith-Lovin and Heise, 1988; MacKinnon, 1994) identifies relations between affective meanings and social events, as follows. People actively create experiences that confirm affective meanings in a given situation. However, when affective meanings are disconfirmed, because people are working at cross-purposes with each other or because of external disturbances, participants may reformulate their definitions of the situation with new affective meanings that fit the experiences that they are having. Emotions arise from comparing self impressions being generated in the situation with the affective meaning of one's identity, signaling to self and others how the confirmation process is going.
A simulation program that implements affect control theory, Interact (Heise, 1978), models the relations between social experience and a definition of the situation (Schneider and Heise, 1995). An analyst using Interact provides a verbal definition of the situation and specifies a first event. Interact looks up words in a cultural database, translating the verbal specifications to EPA profiles defining affective meanings. These EPA profiles then are substituted into empirically-derived impression-formation equations, and the impressions created by the specified event are calculated, along with emotions. An analyst can continue to specify events and observe changes in impressions and emotions. Alternatively, at any point in the sequence, the analyst can compute solutions to problems in which two elements of an event are specified - actor, behavior, or object - and Interact finds the third element that would maximally confirm all affective meanings. For example, the analyst can ask what behavior would optimally confirm the actor and object. Or the analyst can ask what identity the actor should have so that a given behavior toward a given object causes least disconfirmation of meanings. The third element is obtained by a calculus solution minimizing disconfirmations of meanings. Interact does the calculations and reports the resulting EPA profile. Interact also searches the culture database and displays identities or behaviors (whichever is appropriate) whose profiles are similar to the ideal profile.
Interact has been revised and expanded numerous times since development of the first version 25 years ago. Currently the program is available on the World Wide Web as a Java applet and it can simulate social interactions using data from the U.S., Canada, Japan, Germany, and Northern Ireland. During the project period, our goals regarding Interact are as follows.
1. Obtain improved databases for analyzing U.S. social interactions, and incorporate new or expanded databases to permit analyses of social interactions in other cultures.
2. Internationalize the Java applet so that both program texts and simulation inputs-outputs can be presented in indigenous languages, thereby making the program a world-wide resource. (Currently the applet uses English translations in all operations with foreign language databases.)
3. Facilitate the integration of new databases with Interact so that analysts around the world can assemble culture or sub-culture datasets, then use Interact to discover the implications of their numerical measurements, without the intervention of personnel at the central computer server.
4. Improve operations for generating detailed printed reports from the Java applet, possibly including graphic printing of the facial expressions of emotions that Interact generates.
Following are key tasks in this project, personnel involved, and the time parameters for each task. The senior personnel are Professors David Heise, Herman Smith, and Andreas Schneider. This project will involve a subcontract to Professor Schneider at Texas Tech University for the full period of the project, and another subcontract to Professor Smith at the University of Missouri-St. Louis beginning Sep, 2000. (Professor Smith already has an NSF award through August, 2000.)
Complete internationalization and testing of the EPA data-collection applet (D. Heise; software engineering consultants; research assistants) Sep 1999-Aug 2000.
Select stimuli and prepare files for U.S. database replication (D. Heise; H. Smith; A. Schneider; survey research consultants; research assistants) Sep 1999-Aug 2000.
[Select stimuli and prepare files for a Chinese database (H. Smith and research assistants on a prior NSF grant) Mar 1999-Dec 1999.]
Select stimuli and prepare files for expansion of German database (A. Schneider; survey research consultants; research assistants) Jan 2000-Dec 2000.
Enhancement and internationalization of the Interact applet (D. Heise; software engineering consultants; research assistants) Jan 2000-Aug 2002.
Develop the affective-meanings database system to provide more options, more security, integration with Interact, and internationalization (D. Heise; research assistants) Jul 2000-Aug 2002.
Survey of a probability sample of Indianapolis residents (D. Heise; research assistants; Indiana University's Center for Survey Research) Sep 2000-Aug 2001.
Establish field stations for the U.S. database replication (D. Heise; survey research consultants; research assistants) Sep 2000-Dec 2000.
Select stimuli and prepare files for creation of new databases in Africa, South America, Southeast Asia, the former Soviet Union, and/or the Middle East. (D. Heise, H. Smith, A. Schneider; research assistants) Sep 2000-Jun 2001.
Establish field stations for the German database replication (A. Schneider; survey research consultants; research assistants) Jan 2001-Aug 2001.
U.S. data collection and data processing (D. Heise; research assistants) Jan 2001-Dec 2001.
Chinese data collection and data processing (H. Smith; research assistants) Sep 2001-Aug 2002.
German data collection and data processing (A. Schneider; research assistants) Sep 2001-Aug 2002.
Data collection and data processing for Africa, South America, Southeast Asia, the former Soviet Union, and/or the Middle East. (Research assistants) Jul 2001-Aug 2002.
In this project we will engage in both sociological research and software engineering in order to develop procedures for collecting data from remote respondents via the World Wide Web and to internationalize procedures for accessing databases of affective meanings on the World Wide Web. Additionally we will enhance a computer simulation program based on an established social psychological theory in order to facilitate cross-linguistic interpretations of affective-meaning databases via the World Wide Web. This proposed work matches well a suggested Knowledge-Networking topic: "Designing tools for gathering and analyzing data, including new types of tools to collect, share, and manipulate increasingly complex data sets and structures, with underpinning innovations in computing, telecommunications, sophisticated algorithms, and hardware/software systems."
Moreover, as we do methodological research we will be assembling cross-cultural databases of affective meanings to be published on the World Wide Web for use by social scientists - sociologists, psychologists, anthropologists, political scientists, and others - who need to understand affectively-based processes in different cultures.
Heise, D. R. 1978. Computer-Assisted Analysis of Social Action (Chapel Hill, N.C.: Institute for Research in Social Science, 1978).
Heise, D. R. 1979. Understanding Events: Affect and the Construction of Social Action. New York: Cambridge University Press.
Heise, D. R. 1988. Programs INTERACT and ATTITUDE. Dubuque, Iowa: Wm. C. Brown Publishers.
MacKinnon, Neil J. 1994. Symbolic Interactionism as Affect Control. Albany: State University of New York Press.
Osgood, C. H., W. H. May, and M. S. Miron. 1975. Cross-Cultural Universals of Affective Meaning. Urbana: University of Illinois Press.
Romney, A. Kimball, Susan C. Weller, and William H. Batchelder. 1986 Culture as consensus: A theory of culture and informant accuracy, American Anthropologist, 88: 313-338.
Schneider, Andreas. 1997. New Techniques in Cross-Cultural Comparison. Dissertation: Indiana University.
A. Schneider and D. R. Heise. 1995. Simulating Symbolic Interaction. Journal of Mathematical Sociology 20:271-287.
Smith, Herman W., Matsuno, Takanori, and Umino, Michio. 1994. How similar are impression-formation processes among Japanese and Americans? Social Psychology Quarterly, 57: 124-139.
Smith-Lovin, Lynn, and Heise, D. R. 1988. Analyzing Social Interaction: Advances in Affect Control Theory. New York: Gordon and Breach Science Publishers.