Shaping the discussion

Editor’s note: Raúl Pérez is principal moderator and consultant at Utilis/Hispanic Consumer Research, New York. The author would like to thank Michelle Finzel, Diane Harris, Joel Reish, Lauren Woodiwiss and Jay Zaltzman for kindly sharing their experiences with different tabletop setups.

Being constantly on the move from market to market, and from facility to facility, moderators sometimes stumble upon unexpected situations. Though these unexpected situations often alter the smooth progress of projects, a few times they may end up working to a moderator’s advantage.

In the summer of 2008 I moderated focus groups in hotel conference rooms in several small markets that had no focus group facilities. Though every booking manager was asked to provide a rectangular table surface, in one of the markets a circular tabletop was set up instead. Believing that the round shape of the tabletop would not be a major hindrance to the research process, I proceeded to hold the sessions, accepting the unfamiliar arrangement.

To my pleasant satisfaction, participants in the groups in which the circular tabletop was used were very engaged in the discussion and exhibited strong empathy toward each other. While being mindful that no single factor is ever fully responsible for the dynamics of a group, I began wondering to what extent the special setup facilitated an especially safe, welcoming environment. To satisfy my curiosity I conducted a literature search in psychology and business journals on the effect of tabletop shape on group dynamics and then asked a group of fellow moderators about their experiences with circular tabletops.

Direct relationship

Psychology studies have shown a direct relationship between physical distance and verbal production - the nearer a respondent is to an interviewer, the greater the number of words in the answers. Therefore, one way in which tabletop shape may affect group dynamics is the proximity factor. Business management studies have found that participant dissent is more likely when individuals sit around rectangular surfaces than when they sit around circular or similarly-rounded ones.

Although the situations examined by psychologists and business management researchers are different from qualitative marketing research interviews in many respects (e.g., rationale for gathering, individual motivators, etc.), basic commonalities (e.g., sitting around a table, the existence of an interviewer or discussion leader, etc.) are sufficient enough to conclude that proximity and table shape effects should play a significant role in focus groups.

Besides the literature review, I asked a group of fellow moderators about their experiences with circular tabletops. To my surprise my colleagues felt compelled to share their experiences with a variety of tabletop configurations, not just circular ones.

Although these moderators did not believe they could offer an opinion on the effect of a particular tabletop shape on group dynamics, they did have a clear understanding of the effect that different setups have on eye contact, which is a factor closely related to interpersonal dynamics. Moreover, my colleagues were quick to note the impact of different tabletops on clients’ ability to view respondents. As anyone who has observed a focus group will attest, an unobstructed view of respondents greatly facilitates the following of the sessions.

In this article I provide a description of common and not-so-common tabletop shapes used for focus groups. Rather than presenting an exhaustive review of tabletop options for moderators, this article simply attempts to illustrate how tabletop choice, when available, involves a delicate balancing act among moderator, client and respondent needs.

Primarily, the discussion centers on how these surface arrangements may affect group dynamics by altering distance among respondents, proximity of respondents to the moderator, respondents’ ability to make eye contact with each other and the moderator’s ability to visually follow respondents. Secondly, the discussion covers how different tabletop shapes affect clients’ view of respondents. Six different configurations are reviewed.

Rectangular tabletops

The rectangular (or rectangle-like) table is the most common setup available in focus group facilities. It allows the moderator to easily establish eye contact with respondents, although those sitting immediately to the left and right normally tend to fall outside of the moderator’s visual field. Moderators, therefore, usually make an extra effort to visually monitor the pair of respondents sitting to the immediate right or left.

A welcome advantage of rectangular surfaces is that usually there is plenty of workspace for each respondent. Clients sitting right behind the moderator tend to have a nearly unobstructed view of respondents, while those sitting more toward the side of the room tend to experience some visual obstruction caused by respondents or the moderator. Four of the additional tabletop arrangements shown in this article (trapezoid, semicircle, rotated-rectangle and triangle) help overcome these inconvenient viewing conditions.

From a research management perspective, it is easy to mostly worry about the viewing needs of the moderator and clients at the expense of the viewing needs of respondents. Clearly, moderators need effortless eye contact with respondents to lead the session and monitor focus group progress, and clients need an unobstructed view of respondents to follow the sessions they are observing. However, just as importantly, I believe that respondents need effortless visual access to one another to facilitate their engagement in the session. In this regard, I think rectangular tabletops are less than ideal because respondents usually have to make a considerable effort to see other participants sitting on the same side of the table. The two tabletop configurations with curved edges described in this article (circular and semicircular tabletops) help deal with this limitation.

Trapezoid tabletops

Trapezoid tabletops differ from rectangular ones in that the two shortest sides differ in length. Usually the moderator sits on the longer of the two shortest sides of the table. Therefore, one of the main advantages of trapezoid tabletops is that they help alleviate the sight obstruction experienced by clients when rectangular surfaces are used. An additional benefit is that a moderator’s ability to visually follow respondents is somewhat facilitated since the group now appears to open up to a panoramic view. At a very subtle level, the trapezoid shape itself appears to constantly invite clients to engage in the research process. While the trapezoid setup is conducive to better client and moderator viewing of respondents, the adjustment that makes these two benefits possible also leads to two disadvantages: increased interpersonal distance overall and slight impairments to respondents’ ability to make eye contact with each other.

As for the first limitation, if we take a typical rectangular tabletop as an example and increase its width on the moderator’s side, the average distance between each pair of respondents, and between each respondent and the moderator, is increased (see Figure 1). If we were to extrapolate the findings of social psychology studies on interpersonal distance and verbal production to the focus group situation, we might suspect that these increased distances would be a factor limiting the number of words produced by respondents. While number of words per se is not a measure of the quality of group dynamics, it is true that transcripts from very engaged groups yield relatively lengthy participation bits by respondents.

To appreciate the second limitation, let us consider that if most respondents sit along the longer sides of the trapezoid tabletop, then most of them would be facing other respondents in an oblique angle, not straight on as they would in a rectangular setup. This atypical angle, in my opinion, makes it slightly more difficult for respondents to visually engage with each other during the course of the conversation. Also, compared to rectangular tables, the trapezoid arrangement seems to direct respondents’ attention toward the moderator and away from fellow respondents. In other words, in this setup the focus of attention moves toward the moderator and away from respondents.

However plausible, these two slight difficulties should not necessarily affect group dynamics, verbal production or group engagement. For example, focus group interventions (e.g., picture sorts, brand sorts) that throw the focus of attention toward the middle of the table (and away from the moderator) could help address the two types of limitations of trapezoid tabletops just mentioned.

Circular tabletops

Whereas there tends to be more interpersonal space when moderators use a trapezoid table relative to a rectangular shape from which it was derived, there is no more interpersonal distance when moderators employ a circular table of a surface area equivalent to a typical rectangular table (see Figure 2). Furthermore, the average distance between the moderator and respondents is about 17 percent shorter in a circular table than it would be in a rectangular table of equivalent surface area, assuming the sitting arrangement in Figure 2. In terms of respondents’ ability to visually follow each other, respondents are less likely to obstruct the view of those sitting around the table, which is a main limitation of the rectangular setup. Finally, I would argue that a circular setup is a symbolic statement in itself, something that prompts respondents to state to themselves, “We are here in this (group) together,” a desirable idea if we want participants to sense a supporting, safe environment in which to share personal experiences.

Despite their favorable aspects, circular tabletops present important challenges to moderators and clients. If in rectangular setups moderators find it hard to make eye contact with respondents sitting immediately to the left and the right, this issue is dramatically exacerbated when using circular arrangements. A colleague sharing her experience with a circular setup not only found it hard to make eye contact with respondents immediately to the right or left, but also found it difficult to monitor other respondents when paying attention to those immediately to the side.

Needless to say, with circular tables clients are guaranteed an obstructed view of about half of respondents. To some extent, even respondents themselves may be inconvenienced by the circular setup. Some of them would need to significantly alter their sitting position to direct attention to a particular point in the room, for example, a video monitor placed on the wall facing the moderator. In terms of accessible tabletop workspace, there is definitely less of it for each respondent, something which would make exercises requiring multiple materials very difficult.

These disadvantages clearly point toward limiting the use of circular tables to instances in which all of the following conditions apply at the same time: clients are absent; respondent workspace is of minimum importance; and high group cohesion is particularly desirable for the topic at hand. Under these circumstances, circular tables may result in a refreshing experience for the moderator and could facilitate disclosure when dealing with sensitive topics. The literature evaluating the effects of circular and semi-circular tables on business meetings is somewhat consistent with this conclusion: compared to rectangular setups, more-rounded arrangements facilitate a sense of inclusiveness and result in more level relationships among participants.

Semicircular tabletops

In a semicircular table arrangement, a moderator sits at the point corresponding to the center of the circle inscribing the tabletop and respondents sit along the rounded edge. The mirror, as usual, is behind the moderator (Figure 3). Of all arrangements considered in this article, this setup is likely to be the best for client viewing, given the low likelihood that a respondent might block the view of another.

For the moderator, there are two important advantages. One of them, shared by one of my colleagues, is that the potential nuisance of having respondents peek at the discussion guide and other materials is eliminated due to the moderator not having anyone sitting to the immediate right or left. The other advantage is that in this setup the moderator sits relatively equidistant to each respondent, which helps provide a sense of equilibrium to the interactions moderators have with each respondent. This setup, unlike the circular one, gives participants plenty of surface space to work independently with printed materials.

It would be very tempting to assume that semicircular tabletops, given their relation to fully circular shapes, would keep many of the proximity, eye-contact and inclusiveness aspects of circular surfaces. However, upon detailed examination, this table shape has some obvious disadvantages for group dynamics.

First, if circular arrangements facilitate a sense of inclusion and protection that may encourage personal sharing, semicircular tabletops appear to do just the opposite, given that respondents in this latter type of setup are overtly exposed to an unknown audience. Second, when employing semicircular tabletops the average distance between respondents is greater than when using rectangular, trapezoidal or circular tabletops of equivalent surface area, something that would limit interaction. Third, although the view of other respondents is not obstructed by fellow participants, it would appear that the prominent placement of the moderator would magnetize visual attention toward this person, a fact that may affect the quality and quantity of group interaction. It follows that semicircular setups would work best for highly structured situations in which moderator control of the session is paramount; for example, when dealing with highly opinionated experts.

Rotated rectangular setups

The rotated rectangular setup virtually replicates the sitting arrangement of the semicircular setup. As shown in Figure 4, the moderator sits alone in the middle of one of the long sides, with his or her back toward the mirror, and respondents surround the table along the three other sides. While sharing the same strengths and shortcomings of the semicircular setup, there is an additional limitation related to its rectangular shape: Given the likelihood that most respondents would sit along the remaining long side, visual interaction among this subset of respondents would be somewhat limited.

Triangular setups

This rather unusual setup is very useful when conducting focus groups in flexible conference spaces such as those found in hotel meeting rooms. As described by one of my colleagues, three long, narrow tables, one of which is exclusively used by the moderator, are arranged in triangle form (Figure 5). With a camera placed behind and above the moderator, clients located in an adjacent viewing room should be able to have an adequate view of respondents. While smartly solving the problem of how to place a video camera in a hotel room, just as in the semicircular, rotated-rectangular and trapezoidal setups, in the triangular arrangement respondents’ vision is partially projected toward the moderator. However, in my opinion, this setup is a very intelligent compromise between the need to provide client viewing and the need to enable respondent interaction.

Important implications

My goal in this article has been to assist fellow moderators and users of qualitative research in their current appreciation and future use of different focus group table arrangements. If interpersonal proximity, ability to make eye contact and ease of client viewing are factors impinging on the research process, it follows that tabletop shape has important implications for focus groups. Based on what is already known in the scientific and business literatures, and our own experiences as moderators, it is within our reach to continue developing an informed understanding of the procedural impact of different focus group tabletop arrangements on our work.

In a figurative world in which clients and moderators were to be granted the same level of respondent viewing regardless of table shape, the decision on what type of table to use would clearly depend primarily on the type of dynamics we want to facilitate among respondents. No single tabletop configuration is better than another - in the real world that judgment would depend on what we are striving to achieve with our respondents and how well the setup exposes participants to clients and the moderator. Still, it should always be helpful for a moderator to consider the effect of the research environment on the respondent experience.