Usenet Groups, Virtual Community and Consumer Behaviors

Cara Okleshen, University of Nebraska
Sanford Grossbart, University of Nebraska
ABSTRACT - Suggestions that Usenet users represent virtual communities make assumptions about use and perceptions of on-line sites. This study examines on-line use (posting and lurking), perceived membership, perceived information value, and changes in behavior due to Usenet groups. Results indicate that more active participants are more apt to: post, see themselves as members, and value and act on information from the group. Posting, but not lurking, is related to perceived membership. Perceived value of information from the group mediates the relationship between perceived membership and changes in consumer behavior. Findings raise questions about depictions of Usenet groups as virtual communities.
[ to cite ]:
Cara Okleshen and Sanford Grossbart (1998) ,"Usenet Groups, Virtual Community and Consumer Behaviors", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 276-282.

Advances in Consumer Research Volume 25, 1998      Pages 276-282


Cara Okleshen, University of Nebraska

Sanford Grossbart, University of Nebraska


Suggestions that Usenet users represent virtual communities make assumptions about use and perceptions of on-line sites. This study examines on-line use (posting and lurking), perceived membership, perceived information value, and changes in behavior due to Usenet groups. Results indicate that more active participants are more apt to: post, see themselves as members, and value and act on information from the group. Posting, but not lurking, is related to perceived membership. Perceived value of information from the group mediates the relationship between perceived membership and changes in consumer behavior. Findings raise questions about depictions of Usenet groups as virtual communities.


There is no consensus about how consumers view or are influenced by the Internet. The Internet is a new communication medium that combines print and television formats; has many components, including e-mail, the World-Wide-Web, Gopher, Multi-User Dimensions (MUDs), and Usenet groups; and blurs distinctions between mass media and interpersonal communication. Thus, as Maignan (1996) notes, descriptions of the Internet depend on how consumers use on-line resources and capabilities. These descriptions depict the Internet as: a communication tool, a social system, an inforation source, and a place or object of consumption. Maignan’s report that some users view the Internet as a community is consistent with arguments that Usenet users form virtual communities (Barlow 1995; Baym 1993; Fischer, Bristor, and Gainer 1996; Granitz and Ward 1996; Jones 1993; Okleshen, Grossbart, and Kennedy 1996; Rheingold 1993). This study examines some facets of the community characterization of Usenet groups.

Virtual communities are electronic networks of persons that typically lack "real" world, traditional communities’ wide ranges of functions, duration, and and/or depth of interconnectedness and sharing. Virtual communities also vary in breadth, duration, and depth. For example, consumer-related Usenet groups generally are more narrowly focused, lack the histories, and do not foster the depth or life passage rituals of electronic networks, such as Santa Monica’s PEN or the WELL (Shaffer and Anundsen 1993).

Research indicates that consumer-related Usenet groups: offer information, interpretations, and ways to maintain relationships; are highly fluid (users come and go); are interacted with and viewed in varying ways by consumers; and affect some consumers’ purchases (Baym 1993; Granitz and Ward 1996; Maignan 1996). Yet, little is known about how experiences with and views about Usenet groups are related to changes in behaviors. This study examines factors that influence consumers’ perceptions of Usenet group membership, and how perceived membership is related to 1) use of Usenet groups; 2) the value consumers place on group information; and 3) changes in their behaviors.



Usenet groups, commonly called discussion groups, have text-based discussions on specific topics. In some groups a moderator limits comments to the topic, whereas in others discussions are wider ranging. Usenet groups are created daily and possible topics seem unlimited. For example, groups address care of sick children, political cartoons, and running shoes. Each group has frequently asked questions (FAQs) to which newcomers are often directed before participating. These factors suggest dual roles for discussion groups. As Maignan (1996) argues, the Internet blends media’s relational uses and corresponding functions. In social terms, Usenet groups are communication tools, social systems, information sources, and consumption mechanisms. In media terms, these groups: set agendas, convey experiences, and clarify values; facilitate selective social affiliation; transmit information and values; and assist decision making and consumption. These relational uses and functions are reflected in the variables in this study (see Figure 1). Observation frequency and interaction frequency reflect the potential for Usenet groups to focus and encourage discussion. Perceived membership corresponds to the capacity of Usenet groups to permit selective affiliation. Perceived information value reflects the social learning which may come from experience with these groups. Behavioral changes indicate the capacities of Usenet groups to alter consumer behavior. Shaffer and Anundsen (1993) suggest that the disintegration of traditional communities has led to other forms of community, including electronic networks, which are developing along a traditional functional to deep community spectrum. Traditional communities and newer, deep communities are differentiated by the more group focused but less conscious orientation of traditional communities versus the more balanced and individual consciousness of deep communities. A more group focused orientation is reflected by a direct relationship between perceived membership and behavioral change; greater individual consciousness is reflected by a relationship between perceived information value and behaviora change. If Usenet groups act traditionally, their dynamics involve the operation of ties, symbolic place, and necessity (Shaffer and Anundsen 1993) and, arguably, the value expressive and utilitarian influences of normative reference groups (Burnkrant and Cousineau 1975; Kaplan and Miller 1987; Park and Lessig 1977; Ward and Reingen 1990). In these cases, behavioral changes are directly related to perceived membership but not to the conscious value placed on information from the group.

If Usenet groups function on a deeper level, behavioral changes are directly and indirectly related to perceived membership, with the perceived value of information acting as a mediator. The combined direct and indirect relationships reflect the balance of group influence and more conscious evaluation of individual choices in deep communities. A departure from traditional community, without the extremes of deep community, would be evidenced by an indirect, mediated relationship through perceived information value. If perceived information value, but not perceived membership, directly relates to behavioral changes, Usenet groups serve information roles and do not primarily act as normative reference groups. The following sections focus on relationships among these variables.

Interaction, Observation, and Membership

Baym’s (1993) ethnography of community in computer-mediated communication focuses on the (rats) discussion group and its purposes: offering information on soap opera content; interpreting shows; using the common interest in shows to negotiate private issues in public space; and sustaining interpersonal relationships. Baym’s findings suggest that Usenet group discussions initially based on shared interests can lead to social and personal topics and form bases for enduring friendships and acquaintances (Okleshen, Grossbart, and Kennedy 1996; Rheingold 1993).



Usenet members vary in their contributions to discussions. When users act as posters, they contribute to the discussion; when acting as lurkers, users observe without participating. Regular posters give consistency and endurance to the group and lurkers can be drawn into the communal spirit that posters create (Baym 1993; Reid 1993; Rheingold 1993). McLaughlin et al. (1993) describe Usenet groups as fluid, because those who are primarily lurkers come and go without being noticed, but they fail to recognize core posters’ contributions to group continuity. In fact, as Baym (1993) states, "several hundred people write in rats each month, and as many as 41,000 may read it at least once each month" (p. 148). Regular posters also affect lurkers’ views. For example, Granitz and Ward’s (1996) study of the group indicates that a small number of posters affect lurkers’ opinions. This finding parallels reference group literature, which suggests that group influence is positively related to interaction frequency (Leigh and Gabel 1992). Yet, as noted below, in Usenet groups the relationship between interaction frequency and behavior is apt to be mediated by perceived membership.

Although few studies exist, it can be argued that posting and lurking may lead to a sense of membership in a Usenet group (Okleshen, Grossbart, and Kennedy 1996). Posting creates and reflects a sense of continuity and group cohesion because those who post use the medium to build relationships through personal comments, opinions, information, and expertise. Lurking, or observation without discussion, may also contribute to a sense of membership because consumers, when lurking, choose to log into to a specific Usenet group. More frequent posting and lurking are apt to reinforce these impressions and heighten users’ sense of membership.

H1: Interaction frequency ispositively related to perceived membership.

H2: Observation frequency is positively related to perceived membership.

H2 appears logical and consistent with the idea that consumer use influences perceptions of the Internet. Alternatively, it may be argued that the rationale for H2 is weaker than that for H1. Observation reflects lurking without interacting and, thus, may resemble off-line situations in which out group members observe but do not interact with group members. Such observation may be motivated by a lack of involvement, curiosity, desire for information without social interaction, or identification with the group. These factors could undermine support for H2.

Perceived Membership, Perceived Information Value, and Behavioral Changes

Because Usenet groups focus on a specific topic, users generally log-in to gain information and/or discuss subjects related to the topic. These groups may have social value, but this function is secondary to offering information (Okleshen, Grossbart, Kennedy 1996). Thus, to the extent that Usenet groups are influential, their primary role, like that of many reference groups, is to provide information (Assael 1995; Burnkrant and Cousineau 1975; Kaplan and Miller 1987; Park and Lessig 1977; Ward and Reingen 1990). This contention is consistent with Baym (1993) and Maignan’s (1996) reports that users view the Internet as an information source and Granitz and Ward’s (1996) argument that Usenet groups create the potential for computer-mediated word-of-mouth dissemination. A consumer’s evaluation of such information is apt to depend on his/her sense of membership. The stronger a consumer’s perceived membership, the more likely he/she will value Usenet group information. The perceived value of information is a reflection of the distinctiveness and importance of group membership (Leigh and Gabel 1992), which suggests that a consumer is apt to place more value on information from the group if he/she feels a greater sense of group membership.

H3: Perceived membership is positively related to the perceived value of information from the group.

Usenet groups may influence behavior because they serve informational, value expressive, and utilitarian functions by giving marketplace knowledge, enhancing images users project to others, and distributing perceived rewards. Granitz and Ward’s (1996) Usenet coffee group study indicates that purchases are influenced by discussions on recommendations, product use, aspects of consumption (why coffee prices increase), and related products (why inexpensive coffee grinders do not work). Their research provides insights, but its limited focus on one group for two bi-weekly periods may have made it difficult to detect other behavioral changes, including how consumers search for information, where and how they buy, and how they go about buying. If consumers view themselves as members, Usenet groups are more apt to influence these behaviors.

H4: Perceived membership is positively related to behavioral changes.

Although Granitz and Ward report that a Usenet group may affect purchases, the relationship between perceived membership and behavioral changes may also be due to an indirect path through information value. The more consumers value group information, the more likely they are to act on it. This idea is consistent suggestions that consumer groups function as information suppliers more than norm enforcers (Asseal 1995; Burnkrant and Cousineau 1975; Kaplan and Miller 1987; Park and Lessig 1977; Ward and Reingen 1990). Thus, if consumers view information as more valuable, accurate, and reliable, it is more apt to affect their behaviors.

H5: Perceived information value is positively related to behavioral changes.

Hypothesized relationships are depicted in the a priori structural model in Figure 1. For simplicity, measurement models (including errors) are not included. There is no hypothesized path between interaction and observation frequencies because the literature classifies Usenet user behavior as either posting or lurking. There are no direct paths between observation frequency and information value or between interaction frequency and behavioral change, because the virtual community literature characterizes the group as influential in terms of perceptions and behavior. Thus, interaction and observation frequencies are not apt to directly influence the perceived value of information or behaviors. Instead, perceived membership and information value are expected to mediate relationships between observation or interaction frequencies and behavioral changes.


Data were collected via on-line surveys from thirty-seven groups (see Exhibit 1). They focused on tangible goods, specific products, or collections. For example, the Star Wars collectibles group’s topics range from why a Princess Leia figure resembles a man to sales of collections. The survey was posted to each group with a request for participation, an introduction customized to the group, and an e-mail address to direct questions to the researcher. From the two-hundred and thirty-six responses received within a two-week period, there was a usable sample of two-hundred and twenty-four completed surveys.


Items were developed for the five constructs (see Table 1). Three items were initially used to measure observation frequency, interaction frequency, membership perception, and information value. Four items represented behavioral changes. Five point scales were used for interaction and observation frequencies (1=never\5=very often), and for the other constructs (1=strongly disagree\5=strongly agree). The structural model is in Figure 1 (measurement models and errors are omitted).

Confirmatory factor analysis was done to check unidimensionality. Measurement model parameter values, error variances, and reliabilities are in Table 1. The measure, "I look at the group discussion," did not load well on the observation frequency factor (parameter value=.19; error variance=.58; r2=.0006). Items were designed to distinguish between two roles or activities of Usenet users, posting and lurking. The second and third observation frequency measures accomplished this via items such as, "I observe without participating". However, the first item lacked this qualifier, did not distinguish between these tendencies, had poor psychometric properties, and was deleted. Fit indices for the revised model were chi-square=107.18 (df=80, n=224, p=.023); GFI=0.94; and RMR=0.041.

Analysis and Results

Hypotheses were tested with LISREL VIII, using maximum likelihood estimation. Descriptive statistics and the covariance matrix are in Tables 1 and 2. Most item distributions are non-normal, according to the univariate normality test for continuous variables. Of fifteen items, only two have a univariate chi-square below 5.99 (df=2, p=.05), and Mardia’s relative multivariate normality measure indicates that the data are non-normally distributed (chi-square=1336.527, p=0.00). The sample size was too small to use another estimation method that accounted for skewed and kurtotic distriutions; thus, maximum likelihood was chosen because it is fairly robust to small sample sizes and skewness and kurtosis below |2.0| (Schumacker and Lomax 1996). Only the information value construct had skewness or kurtosis greater than that value.

Model Fit

LISREL fit indices indicate that the structural model fits the data fairly well. Joreskog and Sorbom (1989) suggest reporting chi-square, GFI, AGFI, and RMRS. The chi-square goodness of fit statistic for the hypothesized model is 118.20 (df=84, n=224, p=.0083). However, this value, which indicates that the original and reproduced matrix do differ, may be inflated due to non-normally distributed data. The goodness of fit index (GFI) is .93; the goodness of fit adjusted for degrees for freedom (AGFI) is .91; and the standardized root mean square residual (RMRS) is .054.

Steiger (1990) suggests reporting the root means square error of approximation (RMSEA) because it measures lack of fit due to the true difference, versus sampling error, between the original and reproduced covariance matrix. The RMSEA of .043 for the hypothesized model is within the acceptable 0.00 to 0.08 range. For model comparison, Schumacker and Lomax (1996) recommend the normed fit index (NFI) and parsimonious fit index (PFI). The NFI, which tests the hypothesized model versus a baseline null model in which all variables are uncorrelated, is .93. An NFI greater than .90 indicates that the model fits well (it ranges from 0=no fit to 1=perfect fit). The PFI, which accounts for the degrees of freedom given the number of hypothesized paths, is .74 (zero to one scale). However, this low PFI can be explained by two non-significant paths, as discussed below.

Schumacker and Lomax (1996) also suggest checking the expected cross validation index (ECVI), which measures the fit function if the model were cross validated on a separate independent sample. ECVI is .85 (on a zero to infinity scale; a larger value is optimal). Comparable EQS results, which adjust for non-normality, are Satorra-Bentler chi square=152.110 (df=84, n=224, p<.001); NFI=.95; and CFI=.976. It could be argued, as noted by one reviewer, that two of the four behavioral change items deal with the result of being in the discussion group, whereas the other two items deal with group influence. To account for this possible distinction, the former two items were deleted and the revised model was rerun. The results did not substantially differ in terms of models fit, parameter values, or hypothesized results.




Figure 1 contains the structural model with the parameter values. The expected positive relationship between interaction frequency and membership perception (H1) is supported (parameter value=.79, p <.05). However, the expected positive relationship between observation frequency and membership perception (H2) is not present (parameter value=.035, p >.05). The correlation between interaction and observation frequency is -.78 (p <.05). Fifty-eight percent of the variance in membership perception is explained by interaction frequency and observation frequency (error variance=.42).

The positive relationship between membership perception and information value supports H3 (parameter value=.48, p <.05). Membership perception explained twenty-three percent of the variance in information value (error variance=.77). Consumers with a greater sense of group membership place more value on information from the discussion group. Although perceived Usenet group membership is not directly related to changes in behaviors (H4, parameter value=-.15, p >.05), there is a positive relationship between information value and behavioral changes (H5, parameter value=.61, p <.05). The combination of membership perception and information value explains thirty percent of the variance in behavioral change (error variance=.70).


The negative correlation between observation frequency and interaction frequency is consistent with the literature’s suggested distinction between posting and lurking behaviors (Baym 1993; Reid 1993; Rheingold 1993). Lurking behavior, indicated by observation frequency, is unrelated to perceived membership, whereas posting behavior, indicated by interaction frequency, is positively related to perceived membership. These results indicate that consumers who are active participants in Usenet groups feel a stronger sense of membership than less active observers. More active users are more apt to: post to the group, see themselves as members, and value and act on the information from the group. Thus, membership perception does affect behavior, albeit indirectly.





Observation frequency does not affect membership perception, possibly (as noted earlier) because those who primarily observe group discussions may feel like distant outsiders who are watching interaction. Thus, future research is needed to determine if those who are primarily lurkers (in contrast to primarily posters) feel less involved with the topic and other on-line personae, participate out of curiosity or a desire to find specific and limited information, and are less interested in building social ties with other members.

Membership perception is indirectly related to behavioral changes. The strength of a consumer’s membership perception is positively associated with the degree to which he/she values the information in the discussion, which is positively related to the extent of change in his/her behaviors. This result is consistent with the idea that Usenet groups serve informational roles, do not primarily act as normative reference groups, and diverge from traditional community. This divergence is reflected in a less group focused orientation and greater individual consciousness than might be expected in traditional community. The indirect link between membership perception and behavioral changes may be explained by research on social versus informational involvement in Usenet groups. Users may view their access to information, not participants’ social influence, as the primary cause of their behaviors (Assael 1995; Burnkrant and Cousineau 1975; Kaplan and Miller 1987; Park and Lessig 1977; Ward and Reingen 1990).

Social influence is typically strongest in situations in which consumers physically observe and interact with each other (Grossbart, Mittelstaedt, and Murdock 1978). For example, Whyte (1954) discusses how window-unit air conditioner adoption in a neighborhood depends on the appliance’s visibility and social proximity. In a virtual environment, a consumer cannot know a user’s clothing or car brands unless they are openly stated or inferred from clues in the discussion. Thus, influences on behavior may be less socially driven and more informationally driven in a Usenet group, as indicated by the low negative parameter value between membership perception and behavioral changes.

An implication is that although Usenet culture tends to reject advertising and selling (Shirky 1995), perceived membership indirectly affects consumer behaviors via information exchange. These results indicate that more active users place the most value on group information and are most apt use it to modify their behaviors. Research may reveal that opinion leaders, or Usenet topic experts, spread positive or negative information which is valued by others (Granitz and Ward 1996).

Limitations and Extensions

This study can be strengthened and extended in a number of ways. Although the nature of the sample potentially creates self-selection bias, the variance in responses was sufficient to support predicted relationships, and mean responses on behavioral change items were not unduly high. Still, a larger random sample may avoid other undetected potential complications associated witha small convenience sample. The sample composition and size are a function of the medium. Those who primarily lurk come and go while those who primarily post are fewer in number and more permanent (McLaughlin et al. 1993). Because the survey was posted once, the number of lurkers available during that period was limited, and the results may reflect method variance since the data were collected in a single administration. With more research on Internet use, effective methods for obtaining large random samples will be developed and permit evaluation of stability of parameter estimates. Longitudinal analysis will also be useful to verify causality (e.g., the path between perceived membership and perceived information value).

Because of chi-square inflation due to non-normal distribution of the data, the model should be cross-validated on other samples. As more theory is developed about on-line groups, other models should be tested. Tests may reveal similar results or the need for separate models based on a typology of Usenet groups. For example, cross-validation may suggest that less product-based groups which focus on social topics (e.g., religious issues) may affect members’ consumer behaviors in a different way than found in this study. The model may also be extended by examining how simultaneous versus sequential communication modes alter membership perceptions and behaviors. Besides work with other Usenet samples, it will be useful to study social forces in groups in terms of other theories (e.g., diffusion) to examine community differences in virtual environments.

Scales in this study were designed to be unidimensional. In the confirmatory analysis, one measure did not load on the observation frequency factor, and was removed on empirical and theoretical grounds. Such modification should be viewed with skepticism, and more scale validation should be done. The behavioral change scale was designed as a global indicator to examine a general relationship between perceived membership and consumer behavior. As more research is done on Usenet groups, this measure can be refined to include specific behaviors affected by membership and information value. More behavioral measures should also be considered, including actual in-store, off-line, and on-line purchase and search behaviors. Exploring these and the prior issues will offer insights on the nature and dynamics of community in Usenet groups.


Asseal, Henry (1995), Consumer Behavior and Marketing Action. Cincinnati: South-Western College Publishing.

Barlow, John (1995), "Is There a There in Cyberspace?" Utne Reader, 68 (March-April), 53-56.

Baym, Nancy (1993), "The Emergence of Community in Computer-Mediated Communication," in Cybersociety: Computer-Mediated Communication and Community, Steven Jones ed., London: Sage Publications.

Burnkrant, Robert and Alain Cousineau (1975), "Informational and Normative Social Influence in Buyer Behavior," Journal of Consumer Research, 2 (December), 206-215.

Fernback, Jan and Brad Thompson (1995), "Computer-Mediated Communication and the American Collectivity: The Dimensions of Community Within Cyberspace," presented at the Annual Convention of the International Communication Association, Albuquerque, NM (May).

Fischer, Eileen, Julia Bristor, and Brenda Gainer (1996), "Creating or Escaping Community?: An Exploratory Study of Internet Consumers’ Behavior," in Advances in Consumer Research, Kim Corfman and John Lynch eds., Provo, UT: Association for Consumer Research, 23, 178-182.

Granitz, Neil and James Ward (1996, "Virtual Community: A Sociocognitive Analysis," in Advances in Consumer Research, Kim Corfman and John Lynch eds., Provo, UT: Association for Consumer Research, 23, 161-166.

Grossbart, Sanford, Robert Mittelstaedt, and Gene Murdock (1978), "Nearest Neighbor Analysis: Inferring Behavioral Processes From Spatial Patterns," in A. K. Hunt Advances in Consumer Research, Association For Consumer Research, 5, 114-18.

Hoffman, Donna and Thomas Novak (1996), "Marketing in Hypermedia Computer-Mediated Environments: Conceptual Foundations," Journal of Marketing, 60 (July), 50-68.

Jones, Steven (1993), "Understanding Community in the Information Age," in Cybersociety: Computer-Mediated Communication and Community, Steven Jones ed., London: Sage Publications.

Joreskog, Karl and Dag Sorbom (1993), LISREL8 User’s Reference Guide. Chicago: Scientific Software International, Inc.

Kaplan, Martin and Charles Miller (1987), "Group Decision Making and Normative Versus Informational Influence: Effects of Type of Issue and Assigned Decision Role," Journal of Personality and Social Psychology, 53, 306-313.

Leigh, James and Terrance Gabel (1992), "Symbolic Interactionism: Its Effects on Consumer Behavior and Implications for Marketing Strategy," The Journal of Consumer Marketing, 9 (Winter).

Maignan, Isabelle (1996), "Nature and Uses of the Internet: A Qualitative Inquiry," in 1996 AMA Educators’ Proceedings: Enhancing Knowledge Development in Marketing, Cornelia Droge and Roger Calantone eds., Chicago, IL: American Marketing Association, VII, 44-45.

McLaughlin, Margaret, Kerry Osborne, and Christine Smith (1993), "Standards of Conduct on Usenet," in Cybersociety: Computer-Mediated Communication and Community, Steven Jones ed., London: Sage Publications.

Okleshen, Cara, Sanford Grossbart, and Patricia Kennedy (1996), "Seeking Gemeinschaft via Virtual Community," Special Session on Interactive Marketing and Electronic Commerce, American Marketing Association Summer Educators’ Conference.

Park, C. Whan and V. Parker Lessig (1977), "Students and Housewives: Differences in Susceptibility to Reference Group Influence," Journal of Consumer Research, 4 (September), 102-110.

Reid, Elizabeth (1993), "Virtual Worlds: Culture and Imagination," in Cybersociety: Computer-Mediated Communication and Community, Steven Jones ed., London: Sage Publications.

Rheingold, Howard (1993), The Virtual Community: Homesteading on the Electronic Frontier. Massachusetts: Addison Wesley Publishing Co.

Shirky, Clay (1995), Voices from the Net. Emeryville, CA: Ziff-David Press.

Schumacker, Randall, and Richard Lomax (1996), A Beginner’s Guide to Structural Equation Modeling. New Jersey: Lawrence Erlbaum Associates, Inc.

Shaffer, Carolyn and Kristin Anundsen (1993), Creating Community Anywhere. New York: Putnam Publishing Group.

Steiger, J. H. (1990), "Structural Model Evaluation and Modification: An Interval Estimation Approach," Multivariate Behavioral Research, 25, 173-180.

Ward, James and Peter Reingen (1990), "Sociocognitive Analysis of Group Decision Making Among Consumers," Journal of Consumer Research, 17 (December), 245-262.

Whyte, William (1954), "The Web of Word of Mouth," Fortune, (November), 140-143+.