Mechanical Turk or volunteer participant? Comparing the two samples in the study of intelligent personal assistants

2020 ◽  
Vol 57 (1) ◽  
Author(s):  
Irene Lopatovska ◽  
Elena Korshakova
Keyword(s):  
Mechanical Turk ◽  
Two Samples ◽  
Personal Assistants ◽  
Volunteer Participant

Guilty Pleasure? Communicating Sexually Explicit Content on Dating Apps and Disillusionment with App Usage

2019 ◽  
Vol 46 (1) ◽  
pp. 55-85
Author(s):  
Sylvia Niehuis ◽  
Alan Reifman ◽  
Dana A Weiser ◽  
Narissra M Punyanunt-Carter ◽  
Jeanne Flora ◽  
...  
Keyword(s):  
College Students ◽  
Adverse Effects ◽  
Latent Class ◽  
Mechanical Turk ◽  
Sexual Content ◽  
Expectancy Violations ◽  
Dating Apps ◽  
Two Samples ◽  
Sexually Explicit ◽  
Explicit Content

Abstract Some people send or receive sexually explicit images or messages when using dating apps. Seeing unwanted content may produce adverse effects, consistent with expectancy violations theory (EVT), and disillusion some users. To test links between encountering sexually explicit materials and dating app disillusionment (with oneself, with others, and regret over app usage), we surveyed two samples of dating app users. Study 1 (n = 531 college students) focused on Tinder, whereas Study 2 (n = 209 Mechanical Turk workers) examined dating apps broadly. In each study, a latent class analysis sorted users into four groups, based on their dating app engagement with sexual content. Participants who rarely exchanged and did not enjoy sexual content were most regretful, as even one bad experience might have violated their expectations. Contrary to EVT, participants with high enjoyment of explicit materials felt disillusioned with themselves. Participants citing relationship-seeking purposes for app usage were highly disillusioned when heavily involved with explicit content.

scholarly journals Comparing Internet Experiences and Prosociality in Amazon Mechanical Turk and Population-Based Survey Samples

2020 ◽  
Vol 6 ◽  
pp. 237802311988983 ◽  
Author(s):  
Eszter Hargittai ◽  
Aaron Shaw
Keyword(s):  
General Population ◽  
University Students ◽  
National Sample ◽  
Sociodemographic Factors ◽  
Population Based ◽  
Mechanical Turk ◽  
Amazon Mechanical Turk ◽  
Two Samples ◽  
Postal Mail ◽  
Population Based Survey

Given the high cost of traditional survey administration (postal mail, phone) and the limits of convenience samples such as university students, online samples offer a much welcomed alternative. Amazon Mechanical Turk (AMT) has been especially popular among academics for conducting surveys and experiments. Prior research has shown that AMT samples are not representative of the general population along some dimensions, but evidence suggests that these differences may not undermine the validity of AMT research. The authors revisit this comparison by analyzing responses to identical survey questions administered to both a U.S. national sample and AMT participants at the same time. The authors compare the two samples on sociodemographic factors, online experiences, and prosociality. The authors show that the two samples are different not just demographically but also regarding their online behaviors and standard survey measures of prosocial behaviors and attitudes. The authors discuss the implications of these findings for data collected on AMT.

Functions of Imagined Interactions With AI Among Alexa-Users

2021 ◽  
pp. 027623662110053
Author(s):  
Timothy Gaspard ◽  
Phil Madison
Keyword(s):  
Artificial Intelligence ◽  
Psychological Effects ◽  
Mechanical Turk ◽  
Interaction Theory ◽  
Human Interactions ◽  
Personal Assistants ◽  
Amazon's Mechanical Turk ◽  
Machine Communication ◽  
The U.S ◽  
Imagined Interactions

Humans and agents of artificial intelligence (AI) participate in human-machine communication (HMC) more frequently now than ever before – especially in the U.S. Voice powered assistants (VPAs) are widely accessible software agents that enact various social roles, such as personal assistants, and are increasingly packaged with AI-devices to complete simple-tasks, such as sending texts, more efficiently. VPAs are designed to mimic human-human interactions (HHIs) to facilitate more natural human-VPA interactions (HVPAIs). The focus of this study is on the psychological effects of HVPAIs with Amazon’s VPA, Alexa, to identify predictors of frequent Alexa-use through six functions of imagined interactions (IIs) – rehearsal, self-understanding, relational maintenance, conflict linkage, compensation and catharsis. A modified survey of imagined interaction was distributed to 810 self-reported Alexa-users recruited through Amazon’s Mechanical Turk (MTurk). Results suggest that HVPAIs with Alexa impacts the imagination of participants similarly to HHIs, and that use of specific functions of IIs are significant negative predictors of Alexa-use. Moreover, the inclusion of machine-interlocutors as part of imagined interaction theory appears to be compelling as humans and machine interactions evolve in the 21st century.

scholarly journals Punishing campus protesters based on ideology

2019 ◽  
Vol 6 (4) ◽  
pp. 205316801989212
Author(s):  
Jason Giersch
Keyword(s):  
College Students ◽  
Experimental Design ◽  
Political Ideology ◽  
The Other ◽  
Mechanical Turk ◽  
University Campuses ◽  
Political Bias ◽  
Two Samples

Recent protests on university campuses have inspired conservative claims that liberals allow partisanship to color their judgement of disorderly activists. Prior research suggests, however, that both ideologies are prone to political bias. Furthermore, because conservatives are typically more concerned with orderliness and authority, there are theoretical reasons to expect conservatives to respond more forcefully to protests than liberals, especially when those protesters are political opponents. Using an experimental design with two samples, one with Mechanical Turk participants and the other with current college students, the study finds support for the hypotheses that (1) conservatives are more punitive towards protesters than liberals, (2) both ideologies are more likely to punish when protesters are their political opponents, and (3) conservatives’ responses to protesters are more sensitive to their ideology than liberals’. These results support recent studies of the psychology of political ideology and punitiveness.

Electron energy-loss near-edge structure in silicate minerals

1992 ◽  
Vol 50 (2) ◽  
pp. 1258-1259
Author(s):  
D W McComb ◽  
R S Payne ◽  
P L Hansen ◽  
R Brydson
Keyword(s):  
Solid State ◽  
Energy Loss ◽  
Electron Energy ◽  
State Energy ◽  
Local Environment ◽  
Edge Structure ◽  
Silicate Minerals ◽  
Two Samples ◽  
Electron Energy Loss ◽  
Careful Processing

Electron energy-loss near-edge structure (ELNES) is an effective probe of the local geometrical and electronic environment around particular atomic species in the solid state. Energy-loss spectra from several silicate minerals were mostly acquired using a VG HB501 STEM fitted with a parallel detector. Typically a collection angle of ≈8mrad was used, and an energy resolution of ≈0.5eV was achieved.Other authors have indicated that the ELNES of the Si L2,3-edge in α-quartz is dominated by the local environment of the silicon atom i.e. the SiO4 tetrahedron. On this basis, and from results on other minerals, the concept of a coordination fingerprint for certain atoms in minerals has been proposed. The concept is useful in some cases, illustrated here using results from a study of the Al2SiO5 polymorphs (Fig.l). The Al L2,3-edge of kyanite, which contains only 6-coordinate Al, is easily distinguished from andalusite (5- & 6-coordinate Al) and sillimanite (4- & 6-coordinate Al). At the Al K-edge even the latter two samples exhibit differences; with careful processing, the fingerprint for 4-, 5- and 6-coordinate aluminium may be obtained.

Mental representation of economic crisis in Italian and Swiss samples 1The authors would like to thank Maurizio Mistri for his help in the construction of items. This research was supported by a grant from the Italian Murst (ex 40%).

2001 ◽  
Vol 60 (1) ◽  
pp. 11-14 ◽  
Author(s):  
Lucia Savadori ◽  
Eraldo Nicotra ◽  
Rino Rumiati ◽  
Roberto Tamborini
Keyword(s):  
Economic Crisis ◽  
Mental Representation ◽  
Mental Representations ◽  
Social Contexts ◽  
Raw Material ◽  
National Context ◽  
Cluster Procedure ◽  
Total N ◽  
Two Samples ◽  
Important Symptom

The content and structure of mental representation of economic crises were studied and the flexibility of the structure in different social contexts was tested. Italian and Swiss samples (Total N = 98) were compared with respect to their judgments as to how a series of concrete examples of events representing abstract indicators were relevant symptoms of economic crisis. Mental representations were derived using a cluster procedure. Results showed that the relevance of the indicators varied as a function of national context. The growth of unemployment was judged to be by far the most important symptom of an economic crisis but the Swiss sample judged bankruptcies as more symptomatic than Italians who considered inflation, raw material prices and external accounts to be more relevant. A different clustering structure was found for the two samples: the locations of unemployment and gross domestic production indicators were the main differences in representations.

Men Are Hierarchical, Women Are Egalitarian: An Implicit Gender Stereotype

2004 ◽  
Vol 63 (2) ◽  
pp. 107-111 ◽  
Author(s):  
Marianne Schmid Mast
Keyword(s):  
Empirical Evidence ◽  
Implicit Association Test ◽  
Gender Stereotype ◽  
Association Test ◽  
Implicit Association ◽  
Two Samples

The goal of the present study was to provide empirical evidence for the existence of an implicit hierarchy gender stereotype indicating that men are more readily associated with hierarchies and women are more readily associated with egalitarian structures. To measure the implicit hierarchy gender stereotype, the Implicit Association Test (IAT, Greenwald et al., 1998) was used. Two samples of undergraduates (Sample 1: 41 females, 22 males; Sample 2: 35 females, 37 males) completed a newly developed paper-based hierarchy-gender IAT. Results showed that there was an implicit hierarchy gender stereotype: the association between male and hierarchical and between female and egalitarian was stronger than the association between female and hierarchical and between male and egalitarian. Additionally, men had a more pronounced implicit hierarchy gender stereotype than women.

Question Wording Effects in the Assessment of Global Self-Esteem

2000 ◽  
Vol 16 (1) ◽  
pp. 13-19 ◽  
Author(s):  
Martin Dunbar ◽  
Graeme Ford ◽  
Kate Hunt ◽  
Geoff Der
Keyword(s):  
Reading Ability ◽  
Factor Model ◽  
Self Esteem ◽  
Reading Scores ◽  
Poor Reading ◽  
Method Effect ◽  
Two Samples ◽  
Positive Item ◽  
Method Effects ◽  
Best Fit

Summary: Marsh (1996) produced evidence that method effects associated with negatively worded items might be responsible for the results of earlier factor analytic studies that reported finding positive and negative self-esteem factors in the Rosenberg Global self-esteem scale ( Rosenberg, 1965 ). He analyzed data collected from children using a 7-item self-esteem measure. This report details attempts to replicate Marsh 's analysis in data collected from two samples of adults who completed the full 10-item Global Self-Esteem (GSE) scale. The results reported here are similar to those given by Marsh in so much as a correlated uniquenesses model produced a superior fit to the data than the simple one factor model (without correlated uniquenesses) or the often reported two factor (positive and negative self-esteem) model. However, whilst Marsh reported that the best fit was produced by allowing negative item uniquenesses to correlate with each other, the model that produced the best fit to these data was one that contained correlated positive item uniquenesses. Supporting his claim that differential responding to negative and positive self-esteem items reflects a method effect associated with reading ability, Marsh also showed that factors associated with negative and positive items were most distinct among children who had poor reading scores. We report a similar effect among a sample of older adults where the correlation between these factors was compared across two groups who were selected according to their scores on a test of verbal reasoning.

Spearman's Law of Diminishing Returns

2006 ◽  
Vol 27 (4) ◽  
pp. 199-207 ◽  
Author(s):  
Peter Hartmann
Keyword(s):  
Principal Axis ◽  
Age Groups ◽  
Principal Component ◽  
Longitudinal Survey ◽  
Male And Female ◽  
Two Samples ◽  
Diminishing Returns ◽  
The One ◽  
Armed Services ◽  
Principal Axis Factor

Spearman's Law of Diminishing Returns (SLODR) with regard to age was tested in two different databases from the National Longitudinal Survey of Youth. The first database consisted of 6,980 boys and girls aged 12–16 from the 1997 cohort ( NLSY 1997 ). The subjects were tested with a computer-administered adaptive format (CAT) of the Armed Services Vocational Aptitude Battery (ASVAB) consisting of 12 subtests. The second database consisted of 11,448 male and female subjects aged 15–24 from the 1979 cohort ( NLSY 1979 ). These subjects were tested with the older 10-subtest version of the ASVAB. The hypothesis was tested by dividing the sample into Young and Old age groups while keeping IQ fairly constant by a method similar to the one developed and employed by Deary et al. (1996) . The different age groups were subsequently factor-analyzed separately. The eigenvalue of the first principal component (PC1) and the first principal axis factor (PAF1), and the average intercorrelation of the subtests were used as estimates of the g saturation and compared across groups. There were no significant differences in the g saturation across age groups for any of the two samples, thereby pointing to no support for this aspect of Spearman's “Law of Diminishing Returns.”

Structuring Assessments of Psychopathology

2006 ◽  
Vol 27 (2) ◽  
pp. 87-92 ◽  
Author(s):  
Willem K.B. Hofstee ◽  
Dick P.H. Barelds ◽  
Jos M.F. Ten Berge
Keyword(s):  
Principal Components ◽  
Personality Assessment ◽  
Clinical Sample ◽  
Principal Component ◽  
Normal Sample ◽  
Normative Sample ◽  
Assessment Data ◽  
Obsessive Compulsive ◽  
Oblique Rotation ◽  
Two Samples

Hofstee and Ten Berge (2004a) have proposed a new look at personality assessment data, based on a bipolar proportional (-1, .. . 0, .. . +1) scale, a corresponding coefficient of raw-scores likeness L = ΢XY/N, and raw-scores principal component analysis. In a normal sample, the approach resulted in a structure dominated by a first principal component, according to which most people are faintly to mildly socially desirable. We hypothesized that a more differentiated structure would arise in a clinical sample. We analyzed the scores of 775 psychiatric clients on the 132 items of the Dutch Personality Questionnaire (NPV). In comparison to a normative sample (N = 3140), the eigenvalue for the first principal component appeared to be 1.7 times as small, indicating that such clients have less personality (social desirability) in common. Still, the match between the structures in the two samples was excellent after oblique rotation of the loadings. We applied the abridged m-dimensional circumplex design, by which persons are typed by their two highest scores on the principal components, to the scores on the first four principal components. We identified five types: Indignant (1-), Resilient (1-2+), Nervous (1-2-), Obsessive-Compulsive (1-3-), and Introverted (1-4-), covering 40% of the psychiatric sample. Some 26% of the individuals had negligible scores on all type vectors. We discuss the potential and the limitations of our approach in a clinical context.

Sign in / Sign up

Export Citation Format

Share Document