The main focus of this study was to illustrate the applicability of multiple correspondence analysis (MCA) in detecting and representing underlying structures in large datasets used to investigate cognitive ageing. Principal component analysis (PCA) was used to obtain main cognitive dimensions, and MCA was used to detect and explore relationships between cognitive, clinical, physical, and lifestyle variables. Two PCA dimensions were identified (general cognition/executive function and memory), and two MCA dimensions were retained. Poorer cognitive performance was associated with older age, less school years, unhealthier lifestyle indicators, and presence of pathology. The first MCA dimension indicated the clustering of general/executive function and lifestyle indicators and education, while the second association was between memory and clinical parameters and age. The clustering analysis with object scores method was used to identify groups sharing similar characteristics. The weaker cognitive clusters in terms of memory and executive function comprised individuals with characteristics contributing to a higher MCA dimensional mean score (age, less education, and presence of indicators of unhealthier lifestyle habits and/or clinical pathologies). MCA provided a powerful tool to explore complex ageing data, covering multiple and diverse variables, showing if a relationship exists and how variables are related, and offering statistical results that can be seen both analytically and visually. 1. Introduction Analysis of research data requires unique considerations depending on the type of data collected and/or on the main purpose of the research. For instance, while in some cases data is collected in ordinal mode, often it is also obtained in categorized groups. Or, as opposed to the traditional hypothesis testing designed to verify a priori hypotheses about relations between variables, exploratory data analysis is used to identify systematic relations between variables, when there are incomplete a priori expectations as to the nature of those relations. Falling in the latter category, the method correspondence analysis (CA), a (multivariate) descriptive data analytic technique, allows simplifying complex data and provides a detailed description of the data, yielding a simple, yet exhaustive analysis (a review of the development of the correspondence analysis methodology can be found in [1]). Specifically, multiple CA (MCA) allows for the analysis of categorical or categorized variables encompassing more than two categorical variables (whereas “simple”
References
[1]
E. J. Beh, “Simple correspondence analysis: a bibliographic review,” International Statistical Review, vol. 72, no. 2, pp. 257–284, 2004.
[2]
R. A. Johnson and D. W. Wichern, Applied Multivariate Correspondence Analysis, Prentice-Hall, Upper Saddle River, NJ, USA, 6th edition, 2007.
[3]
J. F. Hair, R. L. Tatham, R. E. Anderson, and W. Black, Multivariate Data Analysis, Prentice-Hall, Upper Saddle River, NJ, USA, 5th edition, 1998.
[4]
B. G. Tabachnick and L. S. Fidell, Using Multivariate Statistics, Allyn and Bacon, Boston, Mass, USA, 4th edition, 1996.
[5]
S. E. Clausen, Applied Correspondence Analysis: An Introduction, Quantitative applications in the social science, Sage, Thousand Oaks, Calif, USA, 1998.
[6]
M. Greenacre and T. Hastie, “The geometric interpretation of correspondence analysis,” Journal of the American Statistical Association, vol. 82, no. 398, pp. 437–447, 1987.
[7]
K. M. van Meter, M. A. Schiltz, P. Cibois, and L. Mounier, “Correspondence analysis: a history and French sociological perspective,” in Correspondence Analysis in the Social Sciences: Recent Developments and Applications, M. Greenacre and J. Blasius, Eds., Academic Press, San Diego, Calif, USA, 1994.
[8]
D. Aktürk, S. Gün, and T. Kumuk, “Multiple correspondence analysis technique used in analyzing the categorical data in social sciences,” Journal of Applied Sciences, vol. 7, no. 4, pp. 585–588, 2007.
[9]
N. Sourial, C. Wolfson, B. Zhu et al., “Correspondence analysis is a useful tool to uncover the relationships among categorical variables,” Journal of Clinical Epidemiology, vol. 63, no. 6, pp. 638–646, 2010.
[10]
P. Loslever, “Using multiple correspondence analysis with membership values when the system study yields miscellaneous datasets,” Cybernetics and Systems, vol. 40, no. 7, pp. 633–652, 2009.
[11]
N. C. Santos, P. S. Costa, P. Cunha et al., “Mood is a key determinant of cognitive performance in community-dwelling older adults: a cross-sectional analysis,” Age, 2012.
[12]
A. C. Paulo, A. Sampaio, N. C. Santos et al., “Patterns of cognitive performance in healthy ageing in northern portugal: a cross-sectional analysis,” PLoS One, vol. 6, no. 9, Article ID e24553, 2011.
[13]
K. Anstey and H. Christensen, “Education, activity, health, blood pressure and apolipoprotein E as predictors of cognitive change in old age: a review,” Gerontology, vol. 46, no. 3, pp. 163–177, 2000.
[14]
R. A. Parslow, V. J. Lewis, and R. Nay, “Successful aging: development and testing of a multidimensional model using data from a large sample of older Australians,” Journal of the American Geriatrics Society, vol. 59, no. 11, pp. 2077–2083, 2011.
[15]
T. Archer, A. Fredriksson, E. Schütz, and R. M. Kostrzewa, “Influence of physical exercise on neuroimmunological functioning and health: aging and stress,” Neurotoxicity Research, vol. 20, no. 1, pp. 69–83, 2011.
[16]
F. Marcellini, C. Giuli, R. Papa, C. Gagliardi, M. Malavolta, and E. Mocchegiani, “BMI, life-style and psychological conditions in a sample of elderly Italian men and women,” Journal of Nutrition, Health and Aging, vol. 14, no. 7, pp. 515–522, 2010.
[17]
P. A. Reuter-Lorenz and C. Lustig, “Brain aging: reorganizing discoveries about the aging mind,” Current Opinion in Neurobiology, vol. 15, no. 2, pp. 245–251, 2005.
[18]
T. A. Salthouse, “Selective review of cognitive aging,” Journal of the International Neuropsychological Society, vol. 16, no. 5, pp. 754–760, 2010.
[19]
N. Sourial, C. Wolfson, H. Bergman et al., “A correspondence analysis revealed frailty deficits aggregate and are multidimensional,” Journal of Clinical Epidemiology, vol. 63, no. 6, pp. 647–654, 2010.
[20]
P. S. Costa, N. C. Santos, P. Cunha, J. A. Palha, and N. Sousa, “The use of Bayesian latent class cluster models to classify patterns of cognitive performance in healthy ageing,” PLoS One, 2013.
M. Graffar, “Une methode de classification sociale d'echantillon des populations,” Courrier, vol. 6, pp. 455–459, 1956.
[23]
M. F. Folstein, S. E. Folstein, and P. R. McHugh, “'Mini mental state'. a practical method for grading the cognitive state of patients for the clinician,” Journal of Psychiatric Research, vol. 12, no. 3, pp. 189–198, 1975.
[24]
D. Wechsler, Wechsler Adult Intelligence Scale (WAIS-III), Harcourt Assessment, San Antonio, Tex, USA, 1997.
[25]
H. Buschke, M. Sliwinski, G. Kuslansky, and R. B. Lipton, “Aging, encoding specificity, and memory change in the Double Memory Test,” Journal of the International Neuropsychological Society, vol. 1, no. 5, pp. 483–493, 1995.
[26]
E. Strauss, E. M. S. Sherman, and O. Spreen, A Compendium of Neuropsychological Tests: Administration, Norms and Commentary, Oxford University Press, New York, N, USA, 2006.
[27]
M. Lezak, D. Howieson, and D. Loring, Neuropsychological Assessment, Oxford University Press, New York, NY, USA, 2004.
[28]
World Health Organization, “Waist circumference and waist-hip ratio,” Report of a WHO Expert Consultation, WHO Library Cataloguing-in-Publication Data, Geneva, Switzerland, 2008.
[29]
H. Kim, “Measures of influence in correspondence analysis,” Journal of Statistical Computation and Simulation, vol. 40, pp. 3201–4217, 1992.
[30]
H. Kim, “Influence functions in multiple correspondence analysis,” Korean Journal of Applied Statistics, vol. 7, no. 1, pp. 69–74, 1994.
[31]
W. J. Krzanowski, “Attribute selection in correspondence analysis of incidence matrices,” Applied Statistics, vol. 42, no. 3, pp. 529–541, 1993.
[32]
J. C. Gower, “Discussion of ”a combined approach to contingency table analysis using correspondence analysis and log-linear analysis”,” Applied Statistics, vol. 38, no. 1, pp. 249–292, 1989.
[33]
D. B. Rubin, “Inference and missing data,” Biometrika, vol. 63, no. 3, pp. 581–592, 1976.
[34]
P. Kline, An Easy Guide to Factor Analysis, Routledge, London, UK, 2002.
[35]
R. J. A. Little and D. B. Rubin, Statistical Analysis with Missing Data, Wiley, Hoboken, NJ, USA, 2nd edition, 2002.
[36]
R. B. Cattell, “The meaning and strategic use of factor analysis,” in Handbook of Multivariate Experimental Psychology, R. B. Cattell, Ed., Rand McNally, Chicago, Ill, USA, 1996.
[37]
A. Gifi, Non-Linear Multivariate Analysis,, John Willey & Sons, Chichester, UK, 1996.
[38]
L. Lebart, A. Morineau, and K. M. Warwick, Multivariate Descriptive Statistical Analysis, Wiley, New York, NY, USA, 1984.
[39]
J. Beddington, C. L. Cooper, J. Field et al., “The mental wealth of nations,” Nature, vol. 455, no. 7216, pp. 1057–1060, 2008.
[40]
L. Ferrucci, F. Giallauria, and J. M. Guralnik, “Epidemiology of Aging,” Radiologic Clinics of North America, vol. 46, no. 4, pp. 643–652, 2008.
[41]
N. Yamamoto, G. Yamanaka, E. Takasugi et al., “Lifestyle intervention reversed cognitive function in aged people with diabetes mellitus: two-year follow up,” Diabetes Research and Clinical Practice, vol. 85, no. 3, pp. 343–346, 2009.
[42]
K. G. M. M. Alberti and P. Zimmet, “The metabolic syndrome—a new worldwide definition,” The Lancet, vol. 366, no. 9491, pp. 1059–1062, 2005.
[43]
T. N. Akbaraly, M. Kivimaki, M. J. Shipley et al., “Metabolic syndrome over 10 years and cognitive functioning in late midlife: the Whitehall II study,” Diabetes Care, vol. 33, no. 1, pp. 84–89, 2010.
[44]
M. Kivipelto, E.-L. Helkala, T. H？nninen et al., “Midlife vascular risk factors and late-life mild cognitive impairment: a population-based study,” Neurology, vol. 56, no. 12, pp. 1683–1689, 2001.
[45]
I. Skoog, B. Lernfelt, S. Landahl et al., “15-year longitudinal study of blood pressure and dementia,” Lancet, vol. 347, no. 9009, pp. 1141–1145, 1996.
[46]
M. F. Elias, P. K. Elias, L. M. Sullivan, P. A. Wolf, and R. B. D'Agostino, “Obesity, diabetes and cognitive deficit: the Framingham Heart Study,” Neurobiology of Aging, vol. 26, supplement 1, pp. S11–S16, 2005.
[47]
W. Dai, O. L. Lopez, O. T. Carmichael, J. T. Becker, L. H. Kuller, and H. M. Gach, “Abnormal regional cerebral blood flow in cognitively normal elderly subjects with hypertension,” Stroke, vol. 39, no. 2, pp. 349–354, 2008.
[48]
G. A. Dore, M. F. Elias, M. A. Robbins, M. M. Budge, and P. K. Elias, “Relation between central adiposity and cognitive function in the Maine-Syracuse study: attenuation by physical activity,” Annals of Behavioral Medicine, vol. 35, no. 3, pp. 341–350, 2008.
[49]
J. Gunstad, A. Lhotsky, C. R. Wendell, L. Ferrucci, and A. B. Zonderman, “Longitudinal examination of obesity and cognitive function: results from the baltimore longitudinal study of aging,” Neuroepidemiology, vol. 34, no. 4, pp. 222–229, 2010.
[50]
C. H. van Gool, G. I. J. M. Kempen, B. W. J. H. Penninx, D. J. H. Deeg, A. T. F. Beekman, and J. T. M. van Eijk, “Relationship between changes in depressive symptoms and unhealthy lifestyles in late middle aged and older persons: results from the longitudinal aging study Amsterdam,” Age and Ageing, vol. 32, no. 1, pp. 81–87, 2003.
[51]
A. Dregan, R. Stewart, and M. C. Gulliford, “Cardiovascular risk factors and cognitive decline in adults aged 50 and over: a population-based cohort study,” Age Ageing, vol. 42, no. 3, pp. 338–345, 2013.
[52]
I. Lang, R. B. Wallace, F. A. Huppert, and D. Melzer, “Moderate alcohol consumption in older adults is associated with better cognition and well-being than abstinence,” Age and Ageing, vol. 36, no. 3, pp. 256–261, 2007.
[53]
A. L. Gross, G. W. Rebok, D. E. Ford et al., “Alcohol consumption and domain-specific cognitive function in older adults: Longitudinal data from the johns hopkins precursors study,” Journals of Gerontology B, vol. 66, no. 1, pp. 39–47, 2011.
[54]
G. A. Greendale, C. A. Derby, and P. M. Maki, “Perimenopause and cognition,” Obstetrics & Gynecology Clinics of North America, vol. 38, no. 3, pp. 519–535, 2011.
[55]
B. F. M. Bakker, “A new measure of social status for men and women: the social distance scale,” Netherlands Journal of Social Sciences, vol. 29, no. 2, pp. 113–129, 1993.
[56]
C. Guinot, J. Latreille, D. Malvy et al., “Use of multiple correspondence analysis and cluster analysis to study dietary behaviour: food consumption questionnaire in the SU.VI.MAX. cohort,” European Journal of Epidemiology, vol. 17, no. 6, pp. 505–516, 2001.
[57]
M. L. Burton, E. Greenberger, and C. Hayward, “Mapping the ethnic landscape: personal beliefs about own group's and other groups' traits,” Cross-Cultural Research, vol. 39, no. 4, pp. 351–379, 2005.
[58]
S. M. Cavalcante, L. R. Kerr, S. M. Brignol et al., “Sociodemographic factors and health in a population of children living in families infected with HIV in Fortaleza and Salvador, Brazil,” AIDS Care, vol. 25, no. 5, pp. 550–558, 2012.
[59]
D. B. Panagiotakos and C. Pitsavos, “Interpretation of epidemiological data using multiple correspondence analysis and log-linear models,” Journal of Data Science, vol. 2, pp. 75–86, 2004.
[60]
L. D. Howe, B. Galobardes, A. Matijasevich et al., “Measuring socio-economic position for epidemiological studies in low- and middle-income countries: a methods of measurement in epidemiology paper,” International Journal of Epidemiology, vol. 41, no. 3, pp. 871–886, 2012.
[61]
P. Traissac and Y. Martin-Prevel, “Alternatives to principal components analysis to derive asset-based indices to measure socio-economic position in low- and middle-income countries: the case for multiple correspondence analysis,” International Journal of Epidemiology, vol. 41, no. 4, pp. 1207–1208, 2012.
[62]
P. G. M. Van der Heijden, “Correspondence Analysis of Longitudinal Data,” in Encyclopedia of Biostatistics, P. Armitage and T. Colton, Eds., John Wiley & Sons, Chichester, UK, 2005.
