Este relatório objetiva apresentar as análises introdutória do instrumento “Escala de práticas na prevenção do uso de álcool e outras drogas'', que está em fase de desenvolvimento pelo Centro de Referência em Pesquisa, Intervenção e Avaliação em Álcool e Outras Drogas (http://www.ufjf.br/crepeia/).
O instrumento está sendo validado para população de educadores. O objetivo da pesquisa é oferecer uma medida confiável para avaliação das práticas profissionais de educadores de um curso à distância oferecido pela Secretaria Nacional de Políticas sobre Drogas para aproximadamente 10.000 educadores dos estados de Minas Gerais e Rio de Janeiro.
Durante todo o processo de desenvolvimento, foram utilizadas ferramentas de código-aberto, para facilitar o re-uso das técnicas e procedimentos desenvolvidos. Todo conteúdo do instrumento e de suas etapas estará disponível para o público no repositório (http://github.com/crepeia/ead-senad). Atualmente, o projeto está hospedado no repositório (http://github.com/henriquepgomide/ead-senad).
Neste relatório são apresentadas, análises da escala com base em uma amostra de 136 educadores-tutores do curso. As análises foram conduzidas através da linguagem de programação R usando os pacotes car e psych.
O banco de dados da pesquisa, pode ser obtido no seguinte endereço: (https://github.com/henriquepgomide/ead-senad/blob/master/praticasprofissionais_df.csv).
Os resultados são apresentados por tópicos: caracterização da amostra, avaliação descritiva da escala e análise fatorial exploratória.
library(car) # Function Recode
library(psych) # Function Describe
##
## Attaching package: 'psych'
##
## The following object is masked from 'package:car':
##
## logit
praticasPro <- read.csv("praticasprofissionais_df.csv")
## Summing scales to remove NA's
praticasPro$scaleSum <- rowSums(praticasPro[, 24:66])
## Subset completed observations and consented participation
praticasPro <- subset(praticasPro, subset = praticasPro$termo == "Sim" & praticasPro$estado ==
"Finalizadas" & !is.na(praticasPro$scaleSum))
idade <- as.character(praticasPro$idade)
idade[9] <- "35"
idade[44] <- "29"
idade[69] <- "31"
idade[111] <- 42
praticasPro$age <- as.numeric(gsub("anos(.*)", "", idade))
summary(praticasPro$age) # all
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 25.0 33.0 38.5 39.4 45.2 62.0
by(praticasPro$age, praticasPro$sexo, describe) #by sex
## praticasPro$sexo: Feminino
## vars n mean sd median trimmed mad min max range skew kurtosis se
## 1 1 118 38.91 8.5 38 38.34 8.15 26 62 36 0.57 -0.31 0.78
## --------------------------------------------------------
## praticasPro$sexo: Masculino
## vars n mean sd median trimmed mad min max range skew kurtosis se
## 1 1 18 42.39 7.62 44 42.75 8.15 25 54 29 -0.52 -0.66 1.8
cbind(round(prop.table(table(praticasPro$sexo)), 2))
## [,1]
## Feminino 0.87
## Masculino 0.13
cbind(round(prop.table(table(praticasPro$escolaridade)), 2))
## [,1]
## Ensino Superior Completo 0.09
## Pós-graduação 0.91
cbind(round(prop.table(table(praticasPro$estadocivil)), 2))
## [,1]
## Casado (a) 0.50
## Divorciado (a) 0.13
## Outros 0.08
## Solteiro (a) 0.29
timeWorking <- as.character(praticasPro$tempo.atuacao)
praticasPro$timeWorking <- as.numeric(gsub("anos(.*)", "", timeWorking))
## Warning: NAs introduced by coercion
describe(praticasPro$timeWorking)
## vars n mean sd median trimmed mad min max range skew kurtosis se
## 1 1 123 12.59 8.25 10 11.58 7.41 1 42 41 1 0.63 0.74
cbind(round(prop.table(table(praticasPro$religiao)), 2))
## [,1]
## Católica 0.53
## Espírita 0.17
## Evangélica 0.15
## Outras 0.06
## Sem religião 0.10
cbind(round(prop.table(table(praticasPro$contato.tema)), 2))
## [,1]
## Não 0.37
## Sim 0.63
cbind(round(prop.table(table(praticasPro$lida.com)), 2))
## [,1]
## Não 0.36
## Sim 0.64
cbind(round(prop.table(table(praticasPro$onde.lida.com)), 2))
## [,1]
## Escola 0.51
## Família 0.27
## Outros 0.20
## Serviços de Saúde 0.03
questions <- read.csv("praticasprofissionais_questions.csv", col.names = "Itens",
header = FALSE)
## Warning: cannot open file 'praticasprofissionais_questions.csv': No such
## file or directory
## Error: cannot open the connection
print(questions[1:42, 1], type = "html", justify = "left")
## Error: object 'questions' not found
describe(praticasPro[, 24:66], skew = FALSE)
## vars n mean sd median trimmed mad min max range se
## pp001 1 136 1.91 0.91 2 1.79 1.48 1 4 3 0.08
## pp002 2 136 4.48 0.58 5 4.53 0.00 3 5 2 0.05
## pp003 3 136 2.79 0.94 3 2.80 1.48 1 5 4 0.08
## pp004 4 136 4.35 0.56 4 4.36 0.00 3 5 2 0.05
## pp005 5 136 4.46 0.64 5 4.52 0.00 1 5 4 0.06
## pp006 6 136 3.97 0.84 4 4.05 1.48 2 5 3 0.07
## pp007 7 136 4.54 0.54 5 4.57 0.00 3 5 2 0.05
## pp008 8 136 2.88 0.89 3 2.91 1.48 1 5 4 0.08
## pp009 9 136 4.20 0.74 4 4.28 0.00 2 5 3 0.06
## pp010 10 136 1.77 0.73 2 1.69 1.48 1 4 3 0.06
## pp011 11 136 2.44 0.97 2 2.41 1.48 1 5 4 0.08
## pp012 12 136 3.72 0.80 4 3.78 0.00 1 5 4 0.07
## pp013 13 136 3.24 1.03 3 3.22 1.48 1 5 4 0.09
## pp014 14 136 3.43 0.93 4 3.45 1.48 1 5 4 0.08
## pp015 15 136 3.38 1.03 3 3.39 1.48 1 5 4 0.09
## pp016 16 136 4.14 0.70 4 4.21 0.00 2 5 3 0.06
## pp017 17 136 3.15 0.95 3 3.14 1.48 1 5 4 0.08
## pp018 18 136 3.17 1.06 3 3.20 1.48 1 5 4 0.09
## pp019 19 136 3.13 0.98 3 3.18 1.48 1 5 4 0.08
## pp020 20 136 4.35 0.61 4 4.40 0.00 3 5 2 0.05
## pp021 21 136 2.32 0.98 2 2.25 1.48 1 5 4 0.08
## pp022 22 136 3.76 0.80 4 3.81 0.00 2 5 3 0.07
## pp023 23 136 3.81 0.73 4 3.82 0.00 2 5 3 0.06
## pp024 24 136 3.82 0.87 4 3.91 0.00 1 5 4 0.07
## pp025 25 136 3.51 0.89 4 3.54 0.74 1 5 4 0.08
## pp026 26 136 4.30 0.67 4 4.36 0.00 1 5 4 0.06
## pp027 27 136 1.54 0.67 1 1.46 0.00 1 5 4 0.06
## pp028 28 136 1.96 0.88 2 1.86 1.48 1 5 4 0.08
## pp029 29 136 2.62 1.00 2 2.59 1.48 1 5 4 0.09
## pp030 30 136 3.21 0.93 3 3.22 1.48 1 5 4 0.08
## pp031 31 136 2.35 0.82 2 2.34 1.48 1 4 3 0.07
## pp032 32 136 1.73 0.99 1 1.53 0.00 1 5 4 0.09
## pp033 33 136 3.72 0.80 4 3.74 0.00 1 5 4 0.07
## pp034 34 136 3.82 0.72 4 3.83 0.00 1 5 4 0.06
## pp035 35 136 3.61 0.81 4 3.62 1.48 1 5 4 0.07
## pp036 36 136 3.10 0.89 3 3.12 1.48 1 5 4 0.08
## pp037 37 136 3.61 0.89 4 3.65 0.00 1 5 4 0.08
## pp038 38 136 1.96 0.82 2 1.89 1.48 1 4 3 0.07
## pp039 39 136 1.86 0.83 2 1.75 1.48 1 5 4 0.07
## pp040 40 136 2.47 0.91 2 2.45 1.48 1 5 4 0.08
## pp041 41 136 3.16 0.87 3 3.16 1.48 1 5 4 0.07
## pp042 42 136 2.26 0.87 2 2.19 0.00 1 5 4 0.07
## pp043 43 136 4.30 0.75 4 4.42 1.48 1 5 4 0.06
cor.plot(cor(praticasPro[, 24:66], method = "kendal", use = "complete.obs"),
numbers = TRUE)
alpha(praticasPro[, 24:66])
## Warning: Some items were negatively correlated with total scale and were
## automatically reversed.
##
## Reliability analysis
## Call: alpha(x = praticasPro[, 24:66])
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd
## 0.9 0.91 0.95 0.19 10 0.015 3.7 0.38
##
## lower alpha upper 95% confidence boundaries
## 0.88 0.9 0.93
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se
## pp001- 0.90 0.91 0.95 0.19 9.9 0.015
## pp002 0.90 0.91 0.95 0.19 9.9 0.015
## pp003 0.91 0.91 0.95 0.20 10.3 0.014
## pp004 0.90 0.91 0.95 0.19 9.8 0.015
## pp005 0.90 0.91 0.95 0.19 9.8 0.015
## pp006 0.90 0.91 0.95 0.19 9.8 0.015
## pp007 0.90 0.91 0.95 0.19 9.8 0.015
## pp008- 0.91 0.91 0.95 0.20 10.3 0.014
## pp009 0.90 0.91 0.95 0.19 9.9 0.015
## pp010- 0.90 0.91 0.95 0.19 10.2 0.015
## pp011 0.90 0.91 0.95 0.19 10.0 0.015
## pp012 0.90 0.91 0.95 0.19 9.7 0.015
## pp013 0.90 0.91 0.95 0.19 10.1 0.015
## pp014 0.90 0.91 0.95 0.19 9.7 0.015
## pp015 0.90 0.91 0.95 0.19 9.9 0.015
## pp016 0.90 0.91 0.95 0.19 9.7 0.015
## pp017 0.90 0.91 0.95 0.19 9.8 0.015
## pp018 0.91 0.91 0.95 0.19 10.2 0.015
## pp019 0.90 0.91 0.95 0.19 9.9 0.015
## pp020 0.90 0.91 0.95 0.19 9.6 0.015
## pp021- 0.90 0.91 0.95 0.19 9.9 0.015
## pp022 0.90 0.91 0.95 0.19 9.7 0.015
## pp023 0.90 0.91 0.95 0.19 9.7 0.015
## pp024 0.90 0.91 0.95 0.19 9.7 0.015
## pp025 0.90 0.91 0.95 0.19 9.9 0.015
## pp026 0.90 0.91 0.95 0.19 9.8 0.015
## pp027- 0.90 0.91 0.95 0.19 10.0 0.015
## pp028- 0.90 0.91 0.95 0.19 10.0 0.015
## pp029 0.90 0.91 0.95 0.19 10.0 0.015
## pp030 0.90 0.91 0.95 0.19 9.7 0.015
## pp031- 0.90 0.91 0.95 0.19 9.9 0.015
## pp032- 0.90 0.91 0.95 0.19 10.0 0.015
## pp033 0.90 0.91 0.95 0.19 9.7 0.015
## pp034 0.90 0.91 0.95 0.19 9.7 0.015
## pp035 0.90 0.91 0.95 0.19 9.7 0.015
## pp036 0.91 0.91 0.95 0.20 10.3 0.014
## pp037 0.90 0.91 0.95 0.19 9.8 0.015
## pp038- 0.90 0.91 0.95 0.19 9.7 0.015
## pp039- 0.90 0.91 0.95 0.19 9.8 0.015
## pp040- 0.90 0.91 0.95 0.19 9.5 0.015
## pp041 0.90 0.91 0.95 0.19 10.1 0.015
## pp042- 0.90 0.91 0.95 0.19 9.6 0.015
## pp043 0.90 0.91 0.95 0.19 9.9 0.015
##
## Item statistics
## n r r.cor r.drop mean sd
## pp001- 136 0.42 0.41 0.37 4.1 0.91
## pp002 136 0.44 0.42 0.37 4.5 0.58
## pp003 136 0.19 0.16 0.14 2.8 0.94
## pp004 136 0.50 0.49 0.44 4.3 0.56
## pp005 136 0.52 0.51 0.45 4.5 0.64
## pp006 136 0.49 0.48 0.43 4.0 0.84
## pp007 136 0.53 0.52 0.46 4.5 0.54
## pp008- 136 0.16 0.13 0.10 3.1 0.89
## pp009 136 0.45 0.43 0.38 4.2 0.74
## pp010- 136 0.26 0.24 0.18 4.2 0.73
## pp011 136 0.38 0.36 0.35 2.4 0.97
## pp012 136 0.57 0.57 0.56 3.7 0.80
## pp013 136 0.31 0.29 0.26 3.2 1.03
## pp014 136 0.58 0.58 0.56 3.4 0.93
## pp015 136 0.45 0.43 0.41 3.4 1.03
## pp016 136 0.56 0.56 0.51 4.1 0.70
## pp017 136 0.48 0.48 0.45 3.1 0.95
## pp018 136 0.26 0.24 0.22 3.2 1.06
## pp019 136 0.45 0.45 0.44 3.1 0.98
## pp020 136 0.62 0.62 0.57 4.3 0.61
## pp021- 136 0.41 0.40 0.36 3.7 0.98
## pp022 136 0.56 0.56 0.54 3.8 0.80
## pp023 136 0.60 0.59 0.57 3.8 0.73
## pp024 136 0.60 0.60 0.57 3.8 0.87
## pp025 136 0.47 0.46 0.43 3.5 0.89
## pp026 136 0.49 0.48 0.43 4.3 0.67
## pp027- 136 0.38 0.37 0.31 4.5 0.67
## pp028- 136 0.34 0.32 0.29 4.0 0.88
## pp029 136 0.33 0.32 0.31 2.6 1.00
## pp030 136 0.60 0.60 0.58 3.2 0.93
## pp031- 136 0.45 0.43 0.41 3.6 0.82
## pp032- 136 0.36 0.34 0.30 4.3 0.99
## pp033 136 0.56 0.56 0.53 3.7 0.80
## pp034 136 0.55 0.54 0.52 3.8 0.72
## pp035 136 0.56 0.55 0.53 3.6 0.81
## pp036 136 0.17 0.14 0.13 3.1 0.89
## pp037 136 0.52 0.51 0.49 3.6 0.89
## pp038- 136 0.56 0.55 0.52 4.0 0.82
## pp039- 136 0.53 0.52 0.49 4.1 0.83
## pp040- 136 0.68 0.68 0.65 3.5 0.91
## pp041 136 0.30 0.27 0.24 3.2 0.87
## pp042- 136 0.62 0.62 0.60 3.7 0.87
## pp043 136 0.41 0.40 0.35 4.3 0.75
##
## Non missing response frequency for each item
## 1 2 3 4 5 miss
## pp001 0.38 0.42 0.12 0.08 0.00 0
## pp002 0.00 0.00 0.04 0.43 0.52 0
## pp003 0.07 0.32 0.38 0.20 0.03 0
## pp004 0.00 0.00 0.04 0.57 0.39 0
## pp005 0.01 0.00 0.04 0.44 0.51 0
## pp006 0.00 0.07 0.17 0.49 0.27 0
## pp007 0.00 0.00 0.02 0.42 0.56 0
## pp008 0.07 0.24 0.46 0.21 0.02 0
## pp009 0.00 0.03 0.10 0.51 0.36 0
## pp010 0.38 0.49 0.11 0.02 0.00 0
## pp011 0.13 0.49 0.19 0.17 0.01 0
## pp012 0.01 0.09 0.19 0.60 0.11 0
## pp013 0.03 0.26 0.25 0.38 0.09 0
## pp014 0.01 0.18 0.26 0.46 0.09 0
## pp015 0.03 0.17 0.34 0.32 0.15 0
## pp016 0.00 0.03 0.10 0.58 0.29 0
## pp017 0.03 0.24 0.36 0.31 0.07 0
## pp018 0.07 0.19 0.31 0.35 0.08 0
## pp019 0.06 0.20 0.34 0.36 0.04 0
## pp020 0.00 0.00 0.07 0.51 0.42 0
## pp021 0.18 0.49 0.21 0.10 0.03 0
## pp022 0.00 0.08 0.22 0.55 0.15 0
## pp023 0.00 0.04 0.24 0.57 0.14 0
## pp024 0.01 0.07 0.17 0.56 0.18 0
## pp025 0.01 0.14 0.26 0.50 0.09 0
## pp026 0.01 0.01 0.05 0.54 0.39 0
## pp027 0.53 0.43 0.03 0.01 0.01 0
## pp028 0.33 0.45 0.16 0.05 0.01 0
## pp029 0.10 0.40 0.31 0.14 0.04 0
## pp030 0.03 0.19 0.38 0.33 0.07 0
## pp031 0.14 0.45 0.33 0.08 0.00 0
## pp032 0.51 0.35 0.06 0.04 0.04 0
## pp033 0.01 0.04 0.30 0.51 0.14 0
## pp034 0.01 0.03 0.24 0.60 0.13 0
## pp035 0.01 0.07 0.35 0.46 0.12 0
## pp036 0.04 0.20 0.43 0.29 0.04 0
## pp037 0.01 0.11 0.24 0.51 0.12 0
## pp038 0.31 0.47 0.18 0.04 0.00 0
## pp039 0.35 0.49 0.10 0.04 0.01 0
## pp040 0.12 0.45 0.29 0.12 0.01 0
## pp041 0.01 0.22 0.40 0.32 0.04 0
## pp042 0.16 0.52 0.23 0.07 0.01 0
## pp043 0.01 0.02 0.07 0.47 0.43 0
KMO(praticasPro[, 24:66])
## Kaiser-Meyer-Olkin factor adequacy
## Call: KMO(r = praticasPro[, 24:66])
## Overall MSA = 0.81
## MSA for each item =
## pp001 pp002 pp003 pp004 pp005 pp006 pp007 pp008 pp009 pp010 pp011 pp012
## 0.75 0.82 0.53 0.79 0.80 0.81 0.92 0.59 0.82 0.64 0.78 0.86
## pp013 pp014 pp015 pp016 pp017 pp018 pp019 pp020 pp021 pp022 pp023 pp024
## 0.74 0.80 0.77 0.87 0.73 0.59 0.83 0.90 0.79 0.82 0.86 0.88
## pp025 pp026 pp027 pp028 pp029 pp030 pp031 pp032 pp033 pp034 pp035 pp036
## 0.82 0.82 0.75 0.74 0.73 0.90 0.82 0.71 0.87 0.85 0.87 0.50
## pp037 pp038 pp039 pp040 pp041 pp042 pp043
## 0.83 0.80 0.83 0.82 0.62 0.78 0.78
bartlett.test(praticasPro[, 24:66])
##
## Bartlett test of homogeneity of variances
##
## data: praticasPro[, 24:66]
## Bartlett's K-squared = 300.9, df = 42, p-value < 2.2e-16
fa.parallel(praticasPro[, 24:66], fm = "minres", fa = "both", ylabel = "Eigenvalues") # yields 3 components and 4 factors
## Loading required package: parallel
## Loading required package: MASS
## Parallel analysis suggests that the number of factors = 4 and the number of components = 4
pca <- fa.poly(praticasPro[, 24:66], nfactors = 4, rotate = "oblimin", fm = "minres")
## Loading required package: mvtnorm
## The items do not have an equal number of response alternatives, global set to FALSE
## Warning: NaNs produced
## Warning: Matrix was not positive definite, smoothing was done
## Loading required package: GPArotation
print.psych(pca, digits = 2, cut = 0.3)
## Factor Analysis using method = minres
## Call: fa.poly(x = praticasPro[, 24:66], nfactors = 4, rotate = "oblimin",
## fm = "minres")
## Standardized loadings (pattern matrix) based upon correlation matrix
## MR2 MR1 MR3 MR4 h2 u2 com
## pp001 0.40 0.365 0.63 2.1
## pp002 0.58 0.438 0.56 1.2
## pp003 0.46 0.267 0.73 2.7
## pp004 0.77 0.600 0.40 1.1
## pp005 0.65 0.596 0.40 1.2
## pp006 0.77 0.574 0.43 1.2
## pp007 0.78 0.686 0.31 1.1
## pp008 0.69 0.443 0.56 1.4
## pp009 0.71 0.493 0.51 1.0
## pp010 0.64 0.516 0.48 1.5
## pp011 0.78 0.601 0.40 1.0
## pp012 0.60 0.39 0.649 0.35 1.7
## pp013 0.47 0.41 0.381 0.62 2.5
## pp014 0.77 0.630 0.37 1.0
## pp015 0.32 0.31 0.257 0.74 2.1
## pp016 0.67 0.601 0.40 1.2
## pp017 0.63 0.489 0.51 1.2
## pp018 0.43 0.31 0.317 0.68 2.4
## pp019 0.79 0.591 0.41 1.2
## pp020 0.75 0.714 0.29 1.2
## pp021 0.67 0.545 0.45 1.3
## pp022 0.74 0.592 0.41 1.1
## pp023 0.54 0.483 0.52 1.4
## pp024 0.83 0.712 0.29 1.0
## pp025 0.77 0.550 0.45 1.1
## pp026 0.31 0.42 0.461 0.54 2.8
## pp027 -0.43 0.56 0.633 0.37 2.1
## pp028 0.38 0.241 0.76 1.7
## pp029 0.78 0.620 0.38 1.2
## pp030 0.73 0.605 0.40 1.0
## pp031 -0.43 0.335 0.66 1.8
## pp032 -0.46 0.441 0.56 2.3
## pp033 0.72 0.605 0.39 1.1
## pp034 0.39 0.450 0.55 2.7
## pp035 0.68 0.573 0.43 1.2
## pp036 0.078 0.92 1.5
## pp037 0.56 0.517 0.48 1.5
## pp038 0.63 0.638 0.36 1.5
## pp039 0.53 0.535 0.46 1.7
## pp040 -0.54 0.31 0.695 0.31 2.2
## pp041 0.32 0.32 0.203 0.80 2.5
## pp042 -0.61 0.34 0.691 0.31 1.8
## pp043 0.67 0.508 0.49 1.3
##
## MR2 MR1 MR3 MR4
## SS loadings 6.96 6.97 4.06 3.93
## Proportion Var 0.16 0.16 0.09 0.09
## Cumulative Var 0.16 0.32 0.42 0.51
## Proportion Explained 0.32 0.32 0.19 0.18
## Cumulative Proportion 0.32 0.64 0.82 1.00
##
## With factor correlations of
## MR2 MR1 MR3 MR4
## MR2 1.00 0.30 0.33 -0.19
## MR1 0.30 1.00 0.18 -0.30
## MR3 0.33 0.18 1.00 -0.09
## MR4 -0.19 -0.30 -0.09 1.00
##
## Mean item complexity = 1.6
## Test of the hypothesis that 4 factors are sufficient.
##
## The degrees of freedom for the null model are 903 and the objective function was 101.8 with Chi Square of 12193
## The degrees of freedom for the model are 737 and the objective function was 82.27
##
## The root mean square of the residuals (RMSR) is 0.07
## The df corrected root mean square of the residuals is 0.08
##
## The harmonic number of observations is 136 with the empirical chi square 1241 with prob < 2.9e-28
## The total number of observations was 136 with MLE Chi Square = 9639 with prob < 0
##
## Tucker Lewis Index of factoring reliability = 0.01
## RMSEA index = 0.323 and the 90 % confidence intervals are 0.293 0.303
## BIC = 6018
## Fit based upon off diagonal values = 0.94
## Measures of factor score adequacy
## MR2 MR1 MR3 MR4
## Correlation of scores with factors 1 1 1 1
## Multiple R square of scores with factors 1 1 1 1
## Minimum correlation of possible factor scores 1 1 1 1
fa.diagram(pca)
