Developing an online version of Popnvics PICS for understanding digital relationships (COMM MIL research note)

In my recent research, I've been trying to get more into understanding the psychological dimensions of online relationships - particularly in terms of perceived closeness. I have a great respect for network analysis research that provides insight into the density of social networks, but I feel that this data at times is a bit 'cold' and assumes rather mechanistically and deterministically that we can understand the Gestalt of an interpersonal relationship merely by examining connections and communication frequency. While of course much social media research (including my own) is often more interested in connections over people, I'm trying to come up with novel ways in which to understand perceived closeness - that is, irrespective of one's location in my social media network (i.e. being proximal or more distal to myself as a node) or my communication frequency with another, what is my perception of your closeness to me?*

In looking for metrics, I came across the Perceived Interpersonal Closeness Scale developed by Popovics et al (2003) (.pdf here). They present an assessment of a graphical measure of perceived interpersonal closeness, conceptualized as (taken from the introduction of their publication):

During recent decades researchers and clinicians have shown an interest in studying and measuring closeness-related constructs, some of it in this journal (e.g. Timmerman, Emanuels-Zuurveen, & Emmelkamp, 2000). Sarason, Shearin, Pierce and Sarason (1987) defined the common factor underlying measures of perceived social support as ‘the extent to which an individual is accepted, loved and involved in relationships in which communication is open’ (p. 830). It is suggested that this core factor is interpersonal, socio-emotional closeness, a basic component and function of social support. Closeness, explicitly linked with a closeness–distance model of relationship is a richer and broader term than intimacy (Marks & Floyd, 1996).

Their paper outlines a (rather elegant) argument in defense of their graphical measure of this construct, which essentially asks respondents to place relational partners onto a "bulls-eye" style map, with those individuals placed closer to the center as being more (perceptually) close to the respondent. Their data support and validate this measure, and while I haven't seen it's application much in other areas of research, I find it both inspiring and parsimonious.

So, how do we get this measure online? My first attempt (context, this is a study on brand propinquity hence the "brand name" in parentheses):

Before we move to the next section of the survey, we want to ask you one more question about your relationship with (brand name). Below you will see an image of a target that represents how close you are with different individuals in your daily life. Using this target, we’d like you to tell us how close you currently feel to them by marking the relevant answer from the bulls-eye below. To select an answer, choose the red number in the ring of the bulls-eye that best represents how close you feel to (brand name). You can make your selection by using the drop-down list provided.

Thoughts? I need to work on the presentation quality a bit - looking into coding a Flash-based version in which respondents could 'click and drag' from a list of friends, entities, or any other individual units of analysis and snap them to the bulls-eye graphic. I'm also considering a second layer of interaction where individuals might draw perceptions of the connected-ness between the entities on the map (but that's for another time); in effect, a perceived network density map that could be compared with the more objective measures of network density gleaned from social media data.

Reference:

Popovics, M., Milne, D., & Barrett, P. (2003). The scale of perceived interpersonal closeness (PICS). Clinical Psychology and Psychotherapy, 10, 286-301. [.pdf]

*Of course, I recognize that perceived closeness is likely strongly correlated with actual closeness from a network analysis standpoint, but I also wish to propose for the consideration that these two dimensions do not share a perfect correlation. In fact, the extent to which there is any lack of concordance between these measures becomes an interesting scientific question!

COMM MIL is the Media and Interaction Lab housed in West Virginia University's Department of Communication Studies. More information about the lab can be found here.