Understanding Expertise Search Strategies At Networking Events: An Exploratory Study Using Sociometric Badges - Info and Reading Options

The Internet Archive:

Summary: We aim to collect additional data to complement an earlier study by testing the robustness of one of the core insights that emerged. The original study was an experiment about expertise search strategies at networking events. Like in the earlier study, participants are tasked to search for information that one other person in the room possesses (i.e., their search target) while concurrently being potentially the search target for another participant. The current study aims to investigate the robustness of the insight gained from the above study, namely that those who let others approach them gain more referrals than those who initiate interactions themselves. Detailed characteristics of study: The experiment is set up as an information search game in a setting that emulates a typical networking event. Specifically, participants are tasked to search for a specific other participant (their “search target”) based on short clues (“fun facts”) they are given about this participant at the start of the experiment. To measure how participants navigate the room during this task, we use sociometric badges: small electronic devices participants can wear on their body and which track participant movement during the experiment. Participants (n ≤ 72, as limited by the number of available sociometric badges) will be university students who sign up for the event, which is marketed as a research study about networking. Registration: Upon arrival on the day of the event, students will be given an ID number, a preassigned random number between 0 and 100. This ID number will be clearly displayed on a yellow vest participants will wear for the duration of the experiment. Participants will also be given a sociometric badge that also carries their ID number. A sociometric badge is an electronic device that a participant wear on a lanyard, and which records their movement around the room with the purpose of establishing which other participants they are interacting with at which time. The sociometric badges achieve this by measuring the signal strength (based on ultra-wideband technology) to the badges used by all other participants at shorter than one second intervals. The badges measure the absolute and relative location of participants. The badges do not record conversations. Entry survey: After registration and a short introduction by the research team, participants will be instructed to complete a short entry survey on their mobile phone or other portable device. This entry survey is designed to collect the following information. • Basic personal background information, including gender and programme of study. We will not record any sensitive personal information or person-identifying information. • Some questions designed to measure key personality and cognition variables (e.g., Big Five, altruistic networking, self-monitoring, self-efficacy) that will serve as control variables in our analysis, as well as to compare the sample to our original study, but will not be used to test our hypothesis. • Which other participants they already personally know. Information on prior relationships between participants is necessary for us to assign as a search target in the network experiment someone they do not already know. • Two fun facts about themselves. The two fun facts will form the basis of the search task assigned to the participants. Examples of “fun facts” include: “I eat bread with peanut butter every day for breakfast” and “I am an opera singer in my free time”. Participants are instructed to convey something they believe is highly specific to themselves but not directly visible or traceable to their physical appearance (e.g., gender) or cultural background. The students will be reminded that the fun fact must be something that close friends, but not acquaintances or strangers, will likely know about them. The search task: The two fun facts will form the basis of the search task assigned to all participants: participants will be provided with two fun facts written by another participant and the simple task consists of finding that participant during the event. During the experiment, all participants will be simultaneously searching for someone while potentially being the search target for somebody else. It is at the discretion of participants to provide referrals to other participants during the task. The planned duration for the event is one hour. Assigning search targets will take place directly after all participants have completed the entry survey. Specifically, the survey tool will randomly assign up to five search targets to each participant from the subset of other participants whom they have indicated that do not already know (ensuring all targets are different, avoiding friends of friends, avoiding symmetric pairs, and avoiding any search target to be assigned to more than three participants in any round). The search target’s fun fact will be communicated via the tool used for the entry survey alongside basic instructions for their search: • Only talk in pairs. If a person you want to talk to is already talking to someone else, you will either need to wait until they become available or go and talk to another person. • Track your conversations on the survey tool. After each conversation, please use the tool indicating: 1. The ID number of the person you spoke to 2. If you enquired about the fun facts assigned to you 3. If so, is this person your search target (i.e., wrote the fun facts assigned to you), 4. If not, did this person refer you to someone else? And if so, to whom? • Start each conversation with an icebreaker question. Before enquiring about the “fun facts” assigned to you, you must first both ask your icebreaker question. You find these at the top of the survey page where you are tracking each conversation. • When you indicate you have found the person you were looking for, you will be automatically assigned a new pair of fun facts, so you can start looking for a second person, and then a third, etc. (if time permits). Participants will be instructed to remain seated until we verify that everyone has read the instructions. Then, the floor will open, and participants can start walking around the room. Experimental treatment: We will randomly assign 30% of participants to a treatment condition and the remainder to the control condition. Those in the treatment condition will receive the additional instruction that they have been selected for a “special role” in the game. Specifically, these participants are instructed to always wait for other people to approach them instead of taking the initiative in starting conversations, and to only enquire about their assigned fun facts once they have given the other person the chance to ask about theirs. The only exception is when they believe they have identified their search target, in which case they can approach them. These special instructions are included on the closing page with the assigned search targets and general instructions. Participants are instructed to keep their special role a secret to all other participants. The treatment is designed to create variation in the extent to which individuals initiate conversations, which will allow us to observe if initiation reduces the number of referrals individuals receive during the interactions in the experiment. This is in line with the following hypothesis: Hypothesis: Initiating conversations leads to fewer referrals than letting others approach. Incentive: The incentive for participation will be the chance to win a prize (EarPods). Participants are made aware that the quicker they reach their first search target (in terms of the number of people they speak to), the higher their chance of winning. Those in the treatment condition are additionally made aware that, due to their special treatment, their chance of winning is determined differently: the fewer times they wrongly approach who they believed was the target, the higher their chance of winning. Exit survey: After the event, participants will be asked to complete a short survey about their experiences during the game. This will include questions about the emotions felt during game and the perceived difficulty. We will also ask specific questions to test if the experimental manipulation was effective, such as about the perceived balance between initiating conversations and letting others approach. Hypothesis testing and effect sizes: We will test our above hypothesis by using a two-tailed t-test, where we compare the referral rate (received referral = 1, didn’t receive = 0) between all the interactions in each group. We do so using interaction level data (number of interactions producing a referral, over the total number of interactions capable of producing a referral, within treatment group). Referral rates for both groups (control and treatment) will range between 0 and 1. We calculate detectable effect sizes as follows. In the original study (69 participants, 45 minutes), there were a total of 571 interactions, i.e. 8.27 interactions per person. Scaled up to 72 participants during 60 minutes, this would mean a total of 849 conversations (note that the exact number is unpredictable). However, we must exclude certain types of interactions which could not possibly lead to a referral: those where the participant found their target, and those where the participant did not ask their question. This removes 36.4% of interactions, leaving 363 which may be used for our analysis of referral rates. Scaled up to 72 participants (as limited by the number of sociometric badges we have) and planned duration of the experiment (60 minutes), we expect a total of 505 interactions. Assuming that treated and control participants do not differ in the duration of interactions, this means an estimated 168 interactions for members of the treatment group, and 337 for members of the control group. Interactions between two treated participants should occur rarely or never). With this sample size, we can detect, at a significance level of p< 0.05 and statistical power of 0.8, a minimum effect size of Cohen’s d = 0.27. Since the effect size in the main study was 0.56 (strong effect), we expect to be able to find the same effect in the current study. Further information Research question: When searching for information during networking events, do those who let others approach them gain more referrals than those who initiate interactions themselves? Dependent variables: • Proportion of interactions that produce a referral (excluding interactions that cannot possibly produce a referral). Independent variable: • Initiation of interaction (ego vs. alter) • Treatment vs control Exclusion criteria: We will exclude interactions in which: (a) the participant did not ask about the fun fact, (b) the conversation resulted in finding the search target, or (c) initiation direction was unclear or not logged. Missing data: Missing survey data (e.g., incomplete intake surveys) will be excluded from analyses involving those variables. Missing badge data will be logged, and those participants may be excluded from interaction-level analyses if essential location/interaction data is missing. Data sharing plan: De-identified interaction-level and survey data will be uploaded to OSF after data collection, pending ethical approval and participant consent.

Search for “Understanding Expertise Search Strategies At Networking Events: An Exploratory Study Using Sociometric Badges” downloads:

Visit our Downloads Search page to see if downloads are available.

Find “Understanding Expertise Search Strategies At Networking Events: An Exploratory Study Using Sociometric Badges” in Libraries Near You:

Read or borrow “Understanding Expertise Search Strategies At Networking Events: An Exploratory Study Using Sociometric Badges” from your local library.

• The WorldCat Libraries Catalog: Find a copy of “Understanding Expertise Search Strategies At Networking Events: An Exploratory Study Using Sociometric Badges” at a library near you.