1991
DOI: 10.1037/0022-006x.59.1.27
|View full text |Cite
|
Sign up to set email alerts
|

Analysis of change: Modeling individual growth.

Abstract: Research on change is complicated by problems of measurement and analysis stemming from a conceptualization of change as a series of accumulating increments and decrements. In contrast, individual growth curves depict change as a continuous process underlying individual performance. These two perspectives are reviewed, and some problems with the use of difference scores in the study of change are clarified. Traditional methods are contrasted with growth curve analysis for the purposes of measuring change and s… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
194
0
4

Year Published

1997
1997
2016
2016

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 247 publications
(201 citation statements)
references
References 24 publications
3
194
0
4
Order By: Relevance
“…Description of the neurodevelopmental course of children with HIV infection using detailed statistical analyses of individual patterns of change is another necessary next step in research. 35 Longer term follow-up is also needed to document the developmental outcomes and educational attainments of an increasing number of preschool and school-age survivors who have acquired HIV infection through vertical transmission. 36,37 * Percentages are in parentheses † P values refer to differences between HIV-infected children and each of the other two groups.…”
Section: Discussionmentioning
confidence: 99%
“…Description of the neurodevelopmental course of children with HIV infection using detailed statistical analyses of individual patterns of change is another necessary next step in research. 35 Longer term follow-up is also needed to document the developmental outcomes and educational attainments of an increasing number of preschool and school-age survivors who have acquired HIV infection through vertical transmission. 36,37 * Percentages are in parentheses † P values refer to differences between HIV-infected children and each of the other two groups.…”
Section: Discussionmentioning
confidence: 99%
“…O delineamento longitudinal-misto permite inúmeras possibilidades de análise de dados. Usualmente, há alguma tendência para recorrer a métodos analíticos tradicionais de efeitos fi xos, como a regressão múltipla e análise de variância, que são limitados e restritivos no seu tratamento das medidas repetidas 9,16,24 , sobretudo por utilizarem somente informação completa, o que pode limitar Finalmente, a quarta pergunta tratava da identifi cação da forte variabilidade (i.e., variância estatisticamente signifi cativa) nas trajetórias intraindividuais e nas diferenças entre sujeitos no desenvolvimento da sua força estática manual. Nos três modelos a componente aleatória mostra isso mesmo: variância signifi cativa entre sujeitos (no "Baseline" e na velocidade), e intraindividual (variância residual) (TABELA 3).…”
Section: Discussionunclassified
“…A sua posição pode ser sumariada em três etapas sequenciais: a) o recurso a um modelo teórico bem articulado para interpretar a mudança; b) servido por um delineamento temporal que permita uma visão clara e detalhada do fenômeno em estudo, a que se associa; c) um modelo estatístico sufi cientemente elegante, fl exível, efi ciente e robusto para tratar a informação disponível. Tal como bem referem Bryk e Raudenbush 15 , a Modelação Multinível (MMN) apresenta-se como forte candidata para responder aos desafi os enunciados por Collins 14 , em 2006, face à sua versatilidade em permitir analisar e interpretar, de modo sólido e coerente, a dinâmica da mudança intraindividual e das diferenças interindividuais, bem como a possibilidade de incluir variáveis mediadoras e preditores fi xos e/ou dinâmicos a que se associa um modo adequado de lidar com os "missings by design" gerados pelo delineamento longitudinal-misto 3,14,[16][17] .…”
Section: Palavras-chave: Estudos Longitudinais; Modelação Hierárquicaunclassified
See 1 more Smart Citation
“…In fact, regression to the mean is a phenomenon that has no bearing on the estimation of difference scores in pretest-posttest designs~Maris, 1998!. And contrary to earlier theories about the unreliability of difference scores~e.g., Cronbach & Furby, 1970!, it has been shown that they do not unfairly inflate the estimation of change in persons with lower initial values~Francis, Fletcher, Stuebing, Davidson, & Thompson, 1991! ; moreover, when random assignment is used, they are unaffected by measurement error at pretest or posttest~Maris, 1998!.…”
Section: Family Prevention Counselingmentioning
confidence: 99%