Investigating the Effect of Incidental Affect States on Privacy Behavioral Intention

Abstract

Incidental affects users feel during their online activities may alter their privacy behavioral intentions. We investigate the effect of incidental affect (fear and happy) on privacy behavioral intention. We recruited 330 participants for a within-subjects experiment in three random-controlled user studies. The participants were exposed to three conditions neutral, fear, happy with standardised stimuli videos for incidental affect induction. Fear and happy stimuli films were assigned in random order. The participants’ privacy behavioural intentions (PBI) were measured followed by a Positive and Negative Affect Schedule (PANAS-X) manipulation check on self-reported affect. The PBI and PANAS-X were compared across treatment conditions. We observed a statistically significant difference in PBI and Protection Intention in neutral-fear and neutral-happy comparisons. However across fear and happy conditions, we did not observe any statistically significant change in PBI scores. We offer the first systematic analysis of the impact of incidental affects on Privacy Behavioral Intention (PBI) and its sub-constructs. We are the first to offer a fine-grained analysis of neutral-affect comparisons and interactions offering insights in hitherto unexplained phenomena reported in the field.

Pre-Registration: https://osf.io/c3jy8/.

Appendices

A Within-Subjects Study Variants

In the offline study, participants were invited to complete the user study in a lab environment. All participants were exposed to the two treatment conditions, using a constrained random assignment across - happy and fearful. This assignment assured that the order of stimuli exposed to the participants was fairly balanced. All three studies consist of four parts:

a):

participant registration (including demographics and personality/privacy trait questionnaires),

b):

a control PBI session with a neutral video,

c):

first PBI session with a randomly chosen stimulus video (happy/fearful),

d):

second PBI session with the complementary stimulus video (fearful/happy).

All the three studies were within-subject randomized controlled trials with a random assignment of participants to an order either (1. happy, 2. fearful) or (1. fearful, 2. happy).

Differences. The difference between the studies are:

a):

online participants completed a combination of the pre-task survey and the first phase of the study on the same day while in the offline study the participants completed the registration before stating the main study.

b):

The second difference is no video recording was conducted with the online participants while we video recorded the facial expressions of the offline participants.

c):

The online study uses as assignment a simple uniform random assignment.

d):

The offline study uses a constrained random assignment maintaining a balance between stimulus orders.

B Sample

The sample was combined from three studies with different properties but same methodology: (i) Offline, (ii) AMT, and (iii) Prolific.

Study 1: Offline Lab study. The participants were recruited through flyers and emailing lists within the faculties of Social Sciences, Medical Sciences and Computer Science at the host university. 95 participants from Newcastle University registered online to participate in the study. Of those 95 participants, $N_L = 60$ participants completed the study by physically visiting the lab twice. In terms of ethnicity, 54% of the participants were Caucasian, 26% Asian and 20% were African. In terms of classification of studies, 56.7% of the participants were studying for a postgraduate degree, 37.7% were studying for an undergraduate degree, 3.3% of the participants had secondary school education and the remainder did not report their education background. Table 3 summarizes the descriptive statistics of this sub-sample.

Study 2: Online AMT study. The study was conducted in a series of sessions on Amazon Mechanical Turk. Out of 100 registrations, a total of 70 AMT workers completed both sessions of our study. However, 31 responses were found to be unsuitable, by which retained a sub-sample of $N_M = 39$ observations, described in Table 4.

Study 3: Prolific Academic study. The same experiment was conducted on Prolific with a considerably greater completion rate than in the AMT Study. In the first session of affect study we conducted on Prolific 50 submissions were made; out of which 39 completed the study. A second batch requesting for 200 participants was conducted. 217 completed the first part of the study; 211 returned and completed the study. 15 incorrectly completed surveys were excluded. 235 observations were included in this sub-sample, its descriptives being summarized in Table 5. Table 2 shows the demographics distribution in this sub-sample.

Table 2. Demographics of study 3 on Prolific

C Descriptives

We offer the descriptive statistics for the conditions happy and fearful across the three samples (Lab, MTurk, and Prolific) in Tables 3, 4, and 5, respectively.

Table 3. Descriptives of the lab experiment ($N_{L}=60$)

Table 4. Descriptives of the MTurk experiment ($N_{M}=39$)

Table 5. Descriptives of the Prolific experiment ($N_{M}=226$)