Summary
The paper develops a measurement model for assessing the reliabilities of life event inventories. This method is applied to life event data collected in a longitudinal study of a sample of over 1100 New Zealand children and their families. The findings suggest that the life event inventory was of low-tomodest reliability, with reliability estimates ranging from 0.67 to 0.72. The reliability estimates were applied to the correlations between life event and depression measures collected at two time periods spaced 1 year apart to obtain estimates of the correlations between measures corrected for the effects of test reliability. On the basis of observed and corrected correlations, a non-recursive path model was fitted to the data to examine the effects of measurement error on the coefficients of the model. This analysis showed the presence of quite marked differences in the coefficients of the fitted model. In particular, the model solved assuming that the data were of perfect reliability suggested the presence of reciprocal causation between life event and depression measures, whereas when test reliability was taken into account, the model suggested a uni-directional pattern of causation, in which life event measures influenced depression measures. Nonetheless, both models showed that the predominant direction of association was from life events to depression rather than vice versa. Various theoretical implications of these findings are discussed.
This is a preview of subscription content, access via your institution.
References
Andrews G (1981) A prospective study of life events and psychological symptoms. Psychol Med 11: 795–801
Benjaminsen D (1981) Stressful life events preceding the onset of neurotic depression. Psychol Med 11: 369–378
Brown G, Harris T (1978) The social origins of depression. Tavistock, London
Brown G, Prudo R (1981) Psychiatric disorder in a rural and urban population: 1. Aetiology of depression. Psychol Med 11: 581–599
Cooper B, Sylph J (1973) Life events and the onset of neurotic illness: an investigation in general practice. Psychol Med 3: 421–435
Cronbach L (1951) Coefficient alpha and the internal structure of tests. Psychology 16: 297–334
Fergusson D, Horwood L, Shannon F (1981) Birth placement and childhood disadvantage. Soc Sci Med [E] 15: 315–326
Fergusson D, Horwood L (1984) Life events and depression in women: a structural equation model. Psychol Med 14: 881–889
Finlay-Jones R, Brown G (1981) Types of stressful life events and the onset of anxiety and depression disorders. Psychol Med 11: 803–815
Gersten J, Langner T, Eisenberg L, Simcha-Fagan O (1977) An evaluation of the aetiologic role of stressful life change events in psychiatric disorder. J Health Soc Behav 18: 228–244
Guilford J, Fruchter B (1973) Fundamental statistics in psychology and education. McGraw-Hill, New York
Heise D (1969) Separating reliability and stability in test-retest correlation. Am Soc Rev 34: 93–101
Henderson A, Bryne D, Duncan-Jones P (1981) Neurosis and the social environment. Academic Press, Sydney
Hibbs D (1974) Problems of statistical estimation and causal inference in time-series regression models. In: Costner H (ed) Sociological methodology 1973–1974. Jossey-Bass, San Francisco
Holmes T, Rahe R (1967) The social readjustment rating scale. J Psychosom Res 11: 213–218
Hornstra R, Klassen D (1977) The course of depression. Compr Psychiatry 18: 119–125
Horowitz H, Schaffer D, Hirito D, Wilner N, Levin B (1977) Life event questionnaires for measuring presumptive stress. Psychosom Med 39: 413–431
Joreskog K (1973) A general method for estimating a linear structural equation system. In: Goldberger A, Duncan O (eds) Structural equation models for the social sciences. Seminar, New York
Joreskog K, Sorbom D (1976) LISREL — estimation of linear structural equation systems by maximum likelihood methods. International Educational Services, Chicago
Luskin R (1978) Estimating and interpreting correlations between disturbances and residual path coefficients in non-recursive (and recursive) causal models. Am J Pol Sci 22: 444–474
Masuda M, Holmes T (1967) Magnitude estimations of social readjustments. J Psychosom Res 11: 219–225
Paykel E, Myers J, Dieuelt M, Klevmen G, Lindenthal J, Pepper M (1969) Life events and depression: a controlled study. Arch Gen Psychiatry 21: 753–760
Pilowsky I, Levine S, Boulton D (1969) The classification of depression by numerical taxonomy. Br J Psychiatry 115: 937–945
Pilowsky I, Boulton D (1970) Development of a questionnaire based decision rule for classifying depressed patients. Br J Psychiatry 116: 647–650
Rabkin J, Struening E (1976) Life events, stress and illness. Science 194: 1013–1020
Stewart A, Salt P (1981) Life stress, life styles, depression and illness in adult women. J Pers Soc Psychol 40: 1063–1069
Susser M (1973) Causal thinking in the health sciences. Oxford University Press, London
Tennant C, Bebbington P, Hurry J (1981) The role of life events in depressive illness: is there a substantial causal relation? Psychol Med 11: 379–389
Thomson K, Hendrie H (1972) Environmental stress in primary depressive illness. Arch Gen Psychiatry 26: 130–132
Warheit G (1979) Life events, coping, stress and depressive symptomatology. Am J Psychiatry 136: 502–507
Wheaton B, Muthen B, Alwin D, Summers G (1977) Assessing reliability and stability in panel models. In: Heise D (ed) Sociological methodology. Jossey-Bass, San Francisco
Wiley D, Wiley J (1970) The estimation of measurement error in panel data. Am Soc Rev 35: 112–117
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Fergusson, D.M., Horwood, L.J. The effects of test reliability on relationships between measures of life events and depression. Soc Psychiatry 21, 53–62 (1986). https://doi.org/10.1007/BF00578743
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF00578743
Keywords
- Public Health
- Measurement Error
- Longitudinal Study
- Measurement Model
- Reliability Estimate