Pictorial Naming Specificity across Ages and Cultures:

A Latent Class Analysis of Picture Norms for Younger and Older Americans and Chinese

 

 

Data Files
(HTML)
(XLS)

Overview of Research

Abstract 1

 

Cross-cultural cognition research relies on pictorial stimuli to address how perceptions of common objects vary across population groups.  We examine naming specificity – the degree of detail elicited for object labels2 – across Age (Young-Old) and Culture (American-Chinese).  Segregating picture-item responses into multiple specificity levels allows for formal analysis using latent class techniques and Rost’s (1985) rank-order binomial model.

 

Overall, three naming specificity classes were supported.  Though Age differences were minor, Cultural differences were not: the Chinese showed far greater variation, naming more items both with high and with low specificity than age-matched American counterparts.  Our results differ from prior studies using familiarity and latency measures, and suggest approximately 27% of commonly-used picture items differed across groups, calling their use into question.

 

 

The following links are to downloadable versions of data files for the Pictorial Naming Specificity Project.  Use of Microsoft Internet Explorer is recommended for the HTML files.

 

Specificity_Comparisons_1.htm
Specificity_Comparisons_1.xls

 

 

1  Yoon, C., Feinberg, F., Gutchess, A.H., & Park, D.C. (2004). Pictorial Naming Specificity across Ages and Cultures: A Latent Class Analysis of Picture Norms for Younger and Older Americans and Chinese. Unpublished manuscript. University of Michigan, Ann Arbor.  Under review at Psychological

Methods.  Web Site, containing additional picture-item norms and comparisons across multiple prior studies:  http://agingmind.cns.uiuc.edu/Pict_Norms/

 

2 Snodgrass, J.G., & Vanderwart, M. (1980). A standardized set of 260 pictures: Norms for name agreement, image agreement, familiarity, and visual complexity. Journal of Experimental Psychology: Human Learning and Memory, 6, 174-215.

Contents

Each file contains the following tabs and information:

 

3-Class Comparisons:

Segments (low, medium and high specificity);

Segment sizes (numbers and proportions);

Logistic probabilities;

Segment sizes/numbers for each of four culture-by-age groups (American-Young, American-Old, Chinese-Young, Chinese-Old);

Degree of agreement across culture-by-age groups;

Exact tests of between group differences;

Latent segment (low, medium and high specificity) membership for each of Snodgrass and Vanderwart’s (1980) 260 pictures, for each culture-by-age group.

 

Agreement by Specificity:

Latent segment (low, medium and high specificity) membership for each of Snodgrass and Vanderwart’s (1980) 260 pictures, for each culture-by-age group, ordered by overall degree of specificity, segregated by degree of across-group agreement.

 

Raw Counts and Proportions:

Picture # (1 to 260);

Concept (as per Snodgrass and Vanderwart, 1980);

Culture-by-Age Group # (1 to 4);

Culture (American, Chinese);

Age (Young, Old);

Sample Size (n) [plus: “no answer”, “don’t know object”, “don’t know name”];

Raw Counts for each specificity group ("-2s"; "-1s"; "0s"; "1s"; "2s");

Valid cases;

Proportions for each specificity group ("-2s"; "-1s"; "0s"; "1s"; "2s"),

 

 

For more information, please contact:


Carolyn Yoon, Ph.D. 

Institute for Social Research, Room 5255


426 Thompson St. 
University of Michigan
Ann Arbor, MI, 48104-2321

Telephone: 734-936-2121
Fax: 734-764-3576
E-mail: yoonc@umich.edu

 

 

Investigators

Carolyn Yoon, Ph.D.

University of Michigan Business School 
Institute for Social Research, University of Michigan

Fred Feinberg, Ph.D.

University of Michigan Business School

Angela Hall Gutchess, Ph.D.
Institute for Social Research, University of Michigan

Denise C. Park, Ph.D.
Professor of Psychology

University of Illinois and Beckman Institute