Social psychology applications (02)

Chair: David Hessen, Friday 24th July, 11.30 - 12.50, Boys Smith Room, Fisher Building.

Ling-Wan Chen, Mi-Tao Chen, Department of Education, Kaohsiung Normal University, Taiwan, Hsing-Ming Lee, Department of Early Childhood Care and Education, Shu-Te University, Taiwan. A study on the correlation between gratitude, optimism, death attitudes and sense of meaning in life for university students. Examples from a university in Southern Taiwan. (097)

Lihshing Wang and Shuyan Sun, Division of Educational Studies and Leadership, University of Cincinnati, USA. Disattenuation of correlations due to fallible measurement in meta-analysis: A critique on the debate over “voodoo” correlations in social neuroscience. (234)

Yulia Denisova, Center for Giftedness Research and Development, Moscow City University of Psychology and Education, Russia. Response time registration via verbal reaction recording. (155) ♥

Yury Dodonov, Center for Giftedness Research and Development, Moscow City University of Psychology and Education, Russia. Response time data analysis based on approximating function technique. (157)

ABSTRACTS

A study on the correlation between gratitude, optimism, death attitudes and sense of meaning in life for university students. Examples from a university in Southern Taiwan. (097)
Ling-Wan Chen, Mi-Tao Chen and Hsing-Ming Lee
The study aimed to explore the relation between gratitude, optimism, death attitudes and sense of meaning in life for university students. A questionnaire survey was conducted, targeting 146 students from a university in southern Taiwan, Republic of China; further, four scales were utilized, including “Gratitude Questionnaire-6(GQ-6)”, “Life Orientation Test Revised(LOT-R)” , “Death Attitudes Scale” and “Sense of Meaning in Life Scale”. Data obtained from the survey were sorted by Pearson product-moment correlation, SEM-confirmatory factor analysis and path analysis. The major findings of this research were as follows: The students’ gratitude and optimism were negative to the attitudes toward death, while positive to the sense of meaning in life. Besides, one of students’ attitudes toward death-“escape acceptance” was negative to the sense of meaning in life, whereas another attitude toward death-“approach acceptance” was positive to the sense of meaning in life. Finally, it indicated that 79.9% of the variance in sense of meaning in life was accounted for by the combined effects of gratitude, optimism and death attitudes. 

Disattenuation of correlations due to fallible measurement in meta-analysis: A critique on the debate over “voodoo” correlations in social neuroscience. (234)
Lihshing Wang and Shuyan Sun
Despite decades of psychometric research supporting and advancing Spearman’s (1904) classic work on disattenuation of correlations due to measurement unreliability, its infrequent use in applied research continues to characterize the literature. The inertia to this century-old methodological recommendation is manifested by the recent debate in psychological science that refutes and defends the reported high correlations between mind and brain in social neuroscience research. This study revisits its theoretical framework rooted in the Classical Test Theory paradigm, summarizes milestone studies in the methodological advancement of the inferential properties of disattenuated correlations, examines the trend of reporting practice in published empirical research, and presents Monte Carlo simulation results to demonstrate the damaging effect of failure to correct spuriously low correlations due to fallible measurement. The findings suggest an alarming rate of under-reported disattenuation and lack of change in this trend over the past decade. Re-analysis of the meta-analytic results based on 54 published studies raises serious concerns about their statistically and substantively misleading conclusions. We conclude by refuting the “voodoo correlations” thesis proclaimed by some social neuroscientists (e.g., Vul et al., in press) and recommend a renewed interest in meta-analytic research to adjust for attenuation when reporting the effect sizes of correlations. Implications for future methodogical research in light of recent advances in structural equation modeling and latent variable modeling are discussed.

Response time registration via verbal reaction recording. (155)
Yulia Denisova
In most of the recent studies response time is registered as the time before pressing the appropriate response key. However audio recording of verbal response can be used as another way of response time registration. Response time is calculated as the time before the beginning of stable verbal reaction that can be easily identified using graphical representation of the waveform. In the present study two ways of response time registration are compared for cognitive tasks with two response alternatives. Individual variance of response times is significantly higher for pressing one of the two keys than for verbal responses. Response times tend to differ for left and right hand both in average time and in individual variance. Audio registration allows avoiding measurement errors resulting just from choosing one of the alternatives with the particular hand. For simple tasks with comparatively short response times audio registration allows to avoid inaccuracy resulting from participants’ motor errors. In some cases verbal response registration is the most appropriate way of response time and accuracy fixation (tasks with more than two response alternatives, anagrams solving, etc.). Some examples of such studies are presented; attention is also paid to some technical aspects of audio response time recording. nov, Center for Giftedness Research and Development, Moscow City University of Psychology and Education, Russia. Response time data analysis based on approximating function technique. (157)

Response time data analysis based on approximating function technique. (157)
Yury Dodonov
Average response time and individual variance are mostly calculated in response time data analysis. However these measures are not enough for tasks including similar trials with increasing difficulty. Mathematical technique described in the present study can be used for complex analysis of response time data of such tasks. Average response times are counted separately for each difficulty level. These average times are approximated; individual parameters of the time function are estimated for any given individual. Optimal function for response time approximation is discussed. Using individual approximation parameters individual index expressing the area under the approximation curve is calculated using integral. This index based on simple response time data reflects individual sensibility to the increase of trial difficulty. Response time data collected from visual stimuli recognition tasks are presented as an example of such analysis (235 participants). Individual integral index shows higher correlations with IQ than simple average response time. This mathematical technique is also demonstrated for response time data analysis in emotional sensitivity tasks with increasing difficulty. Universality of the approximation function for this type of tasks is discussed. This method is compared with other ways of response time data approximation (Hick’s law is discussed as one of the examples).