Measuring agreement between two statistics with applications to age standardization

Summary.  We detail a general method for measuring agreement between two statistics. An application is two ratios of directly standardized rates which differ only by the choice of the standard. If the statistics have a high value for the coefficient of agreement then the expected squared difference between the statistics is small relative to the variance of the average of the two statistics, and inferences vary little by changing statistics. The estimation of a coefficient of agreement between two statistics is not straightforward because there is only one pair of observed values, each statistic calculated from the data. We introduce estimators of the coefficient of agreement for two statistics and discuss their use, especially as applied to functions of standardized rates.     

Functional and emotional traits of corporate social media message strategies: Behavioral insights from S&P 500 Facebook data

Drawing from literature regarding public engagement, the Elaboration Likelihood Model (ELM), computer-mediated communication research, and emotion psychology, this study examines the effects of companies’ social media communication strategies on public engagement behaviors as indexed by post likes, shares, and comments. Specifically, it investigates how corporate Facebook posts’ functional traits (functional interactivity and vividness) and emotional traits (emotion presence, valence, and strength) impact public engagement online. Through data mining and computer-assisted sentiment analysis of 33,379 posts from 106 Standard & Poor 500 companies’ Facebook accounts, this study finds a negative effect of functional interactivity but a positive effect of vividness on engagement. It also shows that emotional traits overall yield stronger public engagement outcomes. Two-way interactions between emotional and functional features are also detected. Theoretical and practical implications are discussed.     

Potential for economic development: A quantitative approach

An earlier version of this paper was presented at the Canadian Economic Association Annual Meeting, Hamilton, Ontario, Canada, June 1987.     

Tales of transition: self-narrative and direct scribing in exploring care-leaving

Direct scribing is a technique that transforms oral discourse into written text, basically by typing oral material verbatim . It was developed in practice with youth in child welfare care and adapted for research data collection. The technique flattens the power imbalance in the interview situation and between oral speech and written text, supports self-reflection, democratizes the ownership of written material, is potentially empowering, and is an efficient means to produce high-quality text. The origins in practice are reviewed and a research application is discussed in detail. Thirty youths aged 18/19 from child welfare constructed self-narratives about how they conceptualized and managed the adolescent transition, and collaborated with the researcher in analysing the narrative as a story. Each interview was direct scribed and printed copy given to the participant. Participants were introduced to direct scribing by completing a questionnaire presented on a note-book computer, and some took part in a gender-specific group discussion as a final data-collection event, which was audio-taped and transcribed. The value of direct scribing and self-narrative in this research project and possibilities for other applications are discussed.     

Old and Cold: Older People and Policies Failing to Address Fuel Poverty

Research, funded by the British Gas Help the Aged Partnership and carried out by the Institute of Gerontology, King's College London, explored the multidimensional issues of fuel poverty. A sample of older homeowners and private renters living in England, Scotland and Wales were interviewed in the Spring of 2003 to explore their experiences of keeping their homes warm in the preceding winter. It was found that almost half of the sample for whom full information was available were in fuel poverty. Government schemes failed to address some important issues. Grants were only available to those with “passport benefits”, excluding those who had minimal occupational pensions. Although most respondents had central heating, it was often old and ineffective, yet grants were not available to modernize them. Government schemes did not extend to paying for external and internal insulation for solid wall properties yet many older people live in such property. Several older people lived in rural areas not connected to mains gas. As mains gas currently provides the cheapest fuel, they faced high bills, yet government policies do not address the differential fuel costs in these areas. The culture of many older people in the study contributed to their living in cold homes. They lived frugally and usually turned heating off in daylight hours during winter. It was also a common practice to sleep in an unheated bedroom during winter and to keep the window open at night. Such practices are acknowledged to be unhealthy.     

Gender and the use of pragmatic particles in Indonesian

Studies of male and female conversational styles in several cultures have shown differences in a number of areas, including the use of pragmatic particles. In most studies, women have been shown to behave more cooperatively, and men more competitively. This paper presents a study of two pragmatic particles in Indonesian conversations, kan ;and iya/ya, which cover a range of functions similar to English tag questions and the pragmatic particle you know. Given these similarities, parallel differences in usage might have been expected. However, differences occur much less frequently than has been the case in these other studies, and although the direction of such gender biases as were observed was generally in the same direction as found in other studies, on the whole Indonesian usage by both genders was facilitative. This result is then seen as a reflection of Indonesian cultural values, which place a premium on maintaining the appearance of cooperative behavior.     

Exploring Time Perspective in Greek Young Adults: Validation of the Zimbardo Time Perspective Inventory and Relationships with Mental Health Indicators

In this article we examine the factorial structure of the Greek version of the Zimbardo Time Perspective Inventory (ZTPI; Zimbardo and Boyd in J Personal Soc Psychol 77:1271–1288, 1999), in a sample of 337 university students, using principal axis factoring (PAF) with oblique rotation, and its dimensionality using parallel analysis. Moreover, we evaluate the internal consistency reliability, the convergent validity (through associations with mental health indicators such as dispositional optimism, self-esteem, trait anxiety, depression, and proactive coping), as well as discriminant, and differential validity of this instrument. The results indicated that the ZTPI had a 5-factor structure (past negative, past positive, present fatalistic, present hedonistic, future). Correlational analyses indicated that an aversive view of the past, and a fatalistic attitude toward life were positively and significantly associated with trait anxiety and depression, while they were negatively correlated with self-esteem, proactive coping, and dispositional optimism. Future time perspective was positively associated with proactive coping, whereas a positive attitude toward the past was negatively associated with depression and trait anxiety. Psychometric properties of the five ZTPI scales were satisfactory (Cronbach’s alphas ranging from .710 to .845), thus facilitating the robust investigation of time perspective in Greek speaking populations. However confirmatory factor analyses revealed that the “positive attitude toward the past” dimension might not be a good indicator of time perspective. We discuss the theoretical implications of these findings for future studies of time perspective.     

The Social Contagion of Gambling: How Venue Size Contributes to Player Losses

The Social Facilitation Effect shows performance on many simple tasks is enhanced by crowds of onlookers or co-actors (others performing the same activity). Previous experimental research has shown that Electronic Gaming Machine (EGM) betting behavior is intensified by the belief that others are gambling along with the subject (Rockloff and Dyer, J Gambl Stud 23(1):1–12, 2007). The present study extends these findings by simulating crowds of differing sizes using a fake video-conference along with a live confederate who gambles concurrently with the subjects. Fifty-four male and 81 female subjects aged 18–82 (M = 46.9, SD = 16.7) played a laptop simulated 3-reel EGM using a $20 stake in 3 conditions: (1) alone, (2) in a simulated group of 5 persons plus 1 live confederate, or (3) in a simulated group of 25 persons plus 1 live confederate. The EGM outcomes were rigged with a fixed 20 trial winning sequence followed by an indefinite losing sequence. As hypothesised, gambling intensity, as measured by trials played, speed of betting and final payouts, was progressively greater with larger crowd sizes (P < .05). In contrast, bet-size was slightly lower with larger crowds. The results suggest that gambling venues with more players tend to increase gambling persistence and contribute to greater long term monetary losses.     

On Causal Inferences for Personalized Medicine: How Hidden Causal Assumptions Led to Erroneous Causal Claims About the D-Value

Abstract

Personalized medicine asks if a new treatment will help a particular patient, rather than if it improves the average response in a population. Without a causal model to distinguish these questions, interpretational mistakes arise. These mistakes are seen in an article by Demidenko that recommends the “D-value,” which is the probability that a randomly chosen person from the new-treatment group has a higher value for the outcome than a randomly chosen person from the control-treatment group. The abstract states “The D-value has a clear interpretation as the proportion of patients who get worse after the treatment” with similar assertions appearing later. We show these statements are incorrect because they require assumptions about the potential outcomes which are neither testable in randomized experiments nor plausible in general. The D-value will not equal the proportion of patients who get worse after treatment if (as expected) those outcomes are correlated. Independence of potential outcomes is unrealistic and eliminates any personalized treatment effects; with dependence, the D-value can even imply treatment is better than control even though most patients are harmed by the treatment. Thus, D-values are misleading for personalized medicine. To prevent misunderstandings, we advise incorporating causal models into basic statistics education.