A Course at Coursera

[the-path-to-procrastination]

The association between perceived intelligence and happiness in life

Title

The association between perceived intelligence and happiness in life

 Sarantos Kaptanis, London, UK

Introduction

“Ignorance is bliss”, goes the old saying; similarly “Beati paupers spiritu” (Matthew 5:3), or blessed are the poor in spirit. These sayings are seemingly similar, however there is an important difference: the first one refers to actual intelligence (rather, lack of) and the second one to perceived intelligence, or being humble about one’s ability to perceive the world.

  A recent study of 6870 people in the UK, reported in BBC News and the Journal Psychological Medicine, demonstrated that the first notion may be untrue and that people at the lower end of the intelligence spectrum are less likely to be happy than more intelligent individuals. An earlier study by the University of Edinburgh had found no association between IQ and happiness

  Less is known about the ways perceived intelligence affects happiness. Similarly, perceived intelligence may be linked to greater happiness. The mind is powerful and can have a positive effect on wellbeing.

  Furthermore, male or female sex is associated by other researchers with differences on perceived intelligence and happiness.

  Research Questions

Is perceived level of intelligence associated with happiness

Is the association between perceived level of intelligence and happiness similar for males and females

  Methods

Sample:

Adolescents (7th to 12th grade) who responded in the National Longitudinal Study of Adolescent Health (Add Health) (n=6504)

Add Health is the largest, most comprehensive survey of adolescents ever undertaken in the U.S., funded by the National Institute of Child Health and Human Development (NICHD) with co-funding from 17 other federal agencies.

  Measures

Perceived level of intelligence was assessed with a questionnaire; the question H1SE4 “compared with other people your age, how intelligent are you?” (available answers:  moderately below average, slightly below average, about average, slightly above average, moderately above average, extremely above average, refused, don’t know)

Happiness was assessed with three questions in the questionnaire: H1FS11 “during the past week, you were happy”,  “H1FS15 -You enjoyed life”, “H1FS16 - You felt sad” (available answers:  never or rarely, sometimes, a lot of the time, most of the time or all of the time, refused, don’t know)

Analysis

Analysis was conducted with R: A Language and Environment for Statistical Computing (version 3.0.0, 2013) in the RStudio environment using additional packages Rcmdr, car, ggplot2, multcomp, abind, Hmisc, QuantPsyc, prodlim, xtable

Code is at the end of this post

Results

Univariate

58.4% were male, 51.6% female

55.2% rated their intelligence as above average

2.7% were never happy in the past week; 78.4% were happy a lot of the time or most of the time

3.9% never enjoyed life in the past week; however 80% enjoyed life either a lot of the time or most of the time

Only 52.5% were never sad in the past week; 7% were sad a lot of the time or most of the time

  Bivariate

Those with self rated intelligence above average were more likely to enjoy life (84.0%) compared to those with self rated intelligence below average (75.1%), X-squared = 77.7507, df = 1, p-value < 2.2e-16

Those with self rated intelligence above average were more likely to experience happiness (84.0%) compared to those with self rated intelligence below average (75.1%), X-squared = 77.8, df = 1, p-value < 2.2e-16

Those with self rated intelligence above average were less likely to experience sadness (6.3%) compared to those with self rated intelligence below average (7.9%), X-squared = 6.5, df = 1, p-value = 0.01077

By groups

Importantly, it seems more females are likely to self-rate as being of lower intelligence and be less happy (436, 15%) than males (235, 9.8%)

  Discussion

What might the results mean?

People with self-rated intelligence above average seem to be more likely to enjoy life, be happy and not be sad, disproving the biblical saying “beati pauperes spiritu”.

Strengths

Results are based on a large representative sample of US adolescents

Limitations

Results may not be applicable to individuals outside the United States; self-rating of intelligence was obtained in a single setting and may have reverse causation with mood

Recommended future research

Further research is needed to explore the correlation between intelligence, self reported intelligence and happiness, ideally obtaining self reported intelligence ratings and objective intelligence ratings in multiple instances.

Appendix : Code for the analysis in R

  ## final post

# load data

load("~/Dropbox/coursera-PassionDrivenStatistics/addhealth_pds3.RData")

# univariate

# Replace values with NA using package car

if (!library(car,logical.return = TRUE)) {install.packages("car");library(car)}

addhealth_pds$bio_sex <- Recode(addhealth_pds$bio_sex, 'c(6 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs11 <- Recode(addhealth_pds$h1fs11, 'c(6,8 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs16 <- Recode(addhealth_pds$h1fs16, 'c(6,8 ) =  NA', as.factor.result=TRUE)

# Attach variable labels

var.labels <- attr(addhealth_pds,"var.labels")

# Attach value labels without NA values

addhealth_pds$bio_sex <- factor(addhealth_pds$bio_sex, levels = c(1,2), labels = c("male", "female"))

addhealth_pds$h1se4 <- factor(addhealth_pds$h1se4, levels = c(1,2,3,4,5,6), labels = c("moderately below average", "slightly below average", "about average", "slightly above average", "moderately above average", "extremely above average"))

addhealth_pds$h1fs11 <- factor(addhealth_pds$h1fs11, levels = c(0,1,2,3), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time"))

addhealth_pds$h1fs15 <- factor(addhealth_pds$h1fs15, levels = c(0,1,2,3), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time"))

addhealth_pds$h1fs16 <- factor(addhealth_pds$h1fs16, levels = c(0,1,2,3), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time"))

# Get frequencies as HTML tables with xtable, prodlim, univariateTable

if (!library(prodlim,logical.return = TRUE)) {install.packages("prodlim");library(prodlim)}

if (!library(xtable,logical.return = TRUE)) {install.packages("xtable");library(xtable)}

source("http://192.38.117.59/~tag/download/univariateTable.R")

uni_all <- xtable(summary(univariateTable(~bio_sex + h1se4 + h1fs11 + h1fs15 + h1fs16, data=addhealth_pds)))

print(uni_all, type = "html")

# uni_bio_sex <- xtable(summary(univariateTable(~bio_sex, data=addhealth_pds)))

# uni_h1se4 <- xtable(summary(univariateTable(~h1se4, data=addhealth_pds)))

# uni_h1fs11 <- xtable(summary(univariateTable(~h1fs11, data=addhealth_pds)))

# uni_h1fs15 <- xtable(summary(univariateTable(~h1fs15, data=addhealth_pds)))

# uni_h1fs16 <- xtable(summary(univariateTable(~h1fs16, data=addhealth_pds)))

# print(uni_bio_sex, type = "html")

# print(uni_h1se4, type = "html")

# print(uni_h1fs11, type = "html")

# print(uni_h1fs15, type = "html")

# print(uni_h1fs16, type = "html")

  # Bivariate

# chi square simplified

load("~/Dropbox/coursera-PassionDrivenStatistics/addhealth_pds3.RData")

# create a new data frame

if (!library(car,logical.return = TRUE)) {install.packages("car");library(car)}

simple_bio_sex <- Recode(addhealth_pds$bio_sex, 'c(6 ) =  NA', as.factor.result=TRUE)

simple_h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

simple_h1se4 <- Recode(simple_h1se4, 'c(1,2,3 ) =  0', as.factor.result=TRUE)

simple_h1se4 <- Recode(simple_h1se4, 'c(4,5,6 ) =  1', as.factor.result=TRUE)

simple_h1fs11 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA')

simple_h1fs11 <- Recode(simple_h1fs11, 'c(0,1 ) =  0')

simple_h1fs11 <- Recode(simple_h1fs11, 'c(2,3 ) =  1')

simple_h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA')

simple_h1fs15 <- Recode(simple_h1fs15, 'c(0,1 ) =  0')

simple_h1fs15 <- Recode(simple_h1fs15, 'c(2,3 ) =  1')

simple_h1fs16 <- Recode(addhealth_pds$h1fs16, 'c(6,8 ) =  NA')

simple_h1fs16 <- Recode(simple_h1fs16, 'c(0,1 ) =  0')

simple_h1fs16 <- Recode(simple_h1fs16, 'c(2,3 ) =  1')

simple_bio_sex <- factor(simple_bio_sex, , levels = c(1,2), labels = c("male","female"))

simple_h1se4 <- factor(simple_h1se4, levels = c(0,1), labels = c("below average", "above average"))

simple_h1fs11 <- factor(simple_h1fs11, levels = c(0,1), labels = c("less", "more"))

simple_h1fs15 <- factor(simple_h1fs15, levels = c(0,1), labels = c("less", "more"))

simple_h1fs16 <- factor(simple_h1fs16, levels = c(0,1), labels = c("less", "more"))

my_data <- data.frame(simple_bio_sex,simple_h1se4,simple_h1fs11,simple_h1fs15,simple_h1fs16)

options(digits=3) # only print 3 decimal places

.Table <- xtabs(~simple_h1se4 + simple_h1fs11, data=my_data)

.Table

prop.table(.Table, 1)

.Test <- chisq.test(.Table, correct=FALSE)

.Test

.Table <- xtabs(~simple_h1se4 + simple_h1fs15, data=my_data)

.Table

prop.table(.Table, 1)

.Test <- chisq.test(.Table, correct=FALSE)

.Test

.Table <- xtabs(~simple_h1se4 + simple_h1fs16, data=my_data)

.Table

prop.table(.Table, 1)

.Test <- chisq.test(.Table, correct=FALSE)

.Test

.Table <- xtabs(~simple_h1se4 + simple_h1fs11 + simple_bio_sex, data=my_data)

.Table

prop.table(.Table, 1)

.Test <- chisq.test(.Table, correct=FALSE)

.Test

.Table <- xtabs(~simple_h1se4 + simple_h1fs15 + simple_bio_sex, data=my_data)

.Table

prop.table(.Table, 1)

.Test <- chisq.test(.Table, correct=FALSE)

.Test

.Table <- xtabs(~simple_h1se4 + simple_h1fs16 + simple_bio_sex, data=my_data)

.Table

prop.table(.Table, 1)

.Test <- chisq.test(.Table, correct=FALSE)

.Test

# graph

# Replace values with NA using package car

if (!library(car,logical.return = TRUE)) {install.packages("car");library(car)}

addhealth_pds$bio_sex <- Recode(addhealth_pds$bio_sex, 'c(6 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs11 <- Recode(addhealth_pds$h1fs11, 'c(6,8 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs16 <- Recode(addhealth_pds$h1fs16, 'c(6,8 ) =  NA', as.factor.result=TRUE)

# Attach value labels without NA values

addhealth_pds$bio_sex <- factor(addhealth_pds$bio_sex, levels = c(1,2), labels = c("male", "female"))

addhealth_pds$h1se4 <- factor(addhealth_pds$h1se4, levels = c(1,2,3,4,5,6), labels = c("moderately below average", "slightly below average", "about average", "slightly above average", "moderately above average", "extremely above average"))

addhealth_pds$h1fs11 <- factor(addhealth_pds$h1fs11, levels = c(0,1,2,3), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time"))

addhealth_pds$h1fs15 <- factor(addhealth_pds$h1fs15, levels = c(0,1,2,3), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time"))

addhealth_pds$h1fs16 <- factor(addhealth_pds$h1fs16, levels = c(0,1,2,3), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time"))

if (!library(ggplot2,logical.return = TRUE)) {install.packages("ggplot2");library(ggplot2)}

qplot(h1se4, data=addhealth_pds, geom="bar", fill=h1fs11, xlab = "Perceived intelligence") + labs(fill="during the past week, you were happy")

[the-path-to-procrastination]

Can't get HTML tables to work on Tumblr :(

Sad. Anyway, will end up publishing tables as images!

[the-path-to-procrastination]

Assignment 9 (Moderation)

As things increase in complexity, R demonstrates the superiority of its vast userbase and open source model. There are packages that deal solely with moderation (and mediation for that matter), like QuantPsyc. Despite the fact that testing for moderation in other analyses than this sounds sexier, life is short and time is swift.

I will test moderation of the effect of H1FS15 -You enjoyed life  on  H1FS16 - You felt sad, by H1FS11‐ during the past week, you were happy.

  The result confirm a statistically significant effect of the moderation, but only at the level of p=0.0065. Considering the other effects, this is only minimal...

  > lm.mod1 <- moderate.lm(addhealth_pds$h1fs15, addhealth_pds$h1fs11, addhealth_pds$h1fs16, addhealth_pds, mc = FALSE)

The following objects are masked from data (position 3):

    aid, bio_sex, h1ed7, h1ed9, h1fp7, h1fp8, h1fp9m, h1fs1, h1fs10,

    h1fs11, h1fs12, h1fs13, h1fs14, h1fs15, h1fs16, h1fs17, h1fs18,

    h1fs19, h1fs2, h1fs3, h1fs4, h1fs6, h1fs7, h1fs8, h1fs9, h1gi20,

    h1gi21, h1gi4, h1gi6a, h1gi6b, h1gi6c, h1gi6d, h1jo10, h1nr1,

    h1nr2, h1nr3, h1nr5, h1nr6, h1pa1, h1pa2, h1pa3, h1pa4, h1pa5,

    h1pa6, h1pa7, h1re1, h1re2, h1re3, h1re4, h1re5, h1re6, h1re7,

    h1rp1, h1rp2, h1rp3, h1rp4, h1rp5, h1rp6, h1rr1, h1se1, h1se2,

    h1se3, h1se4, h1su1, h1su2

The following objects are masked from data (position 4):

    aid, bio_sex, h1ed7, h1ed9, h1fp7, h1fp8, h1fp9m, h1fs1, h1fs10,

    h1fs11, h1fs12, h1fs13, h1fs14, h1fs15, h1fs16, h1fs17, h1fs18,

    h1fs19, h1fs2, h1fs3, h1fs4, h1fs6, h1fs7, h1fs8, h1fs9, h1gi20,

    h1gi21, h1gi4, h1gi6a, h1gi6b, h1gi6c, h1gi6d, h1jo10, h1nr1,

    h1nr2, h1nr3, h1nr5, h1nr6, h1pa1, h1pa2, h1pa3, h1pa4, h1pa5,

    h1pa6, h1pa7, h1re1, h1re2, h1re3, h1re4, h1re5, h1re6, h1re7,

    h1rp1, h1rp2, h1rp3, h1rp4, h1rp5, h1rp6, h1rr1, h1se1, h1se2,

    h1se3, h1se4, h1su1, h1su2

> summary(lm.mod1)

Call:

lm(formula = y ~ mcx * mcz, na.action = na.omit)

Residuals:

            Min              1Q          Median              3Q             Max 

-1.380675990106 -0.459108167610 -0.313833369548  0.395617034327  2.686166630452 

Coefficients:

                     Estimate        Std. Error   t value   Pr(>|t|)    

(Intercept)  0.55294469645634  0.00885787947522  62.42405 < 2.22e-16 ***

mcx         -0.14865855640759  0.01127539086289 -13.18434 < 2.22e-16 ***

mcz         -0.16717556993606  0.01178241391825 -14.18857 < 2.22e-16 ***

mcx:mcz      0.02900130795672  0.01065908626369   2.72081  0.0065297 ** 

---

Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.634392877597 on 6481 degrees of freedom

  (19 observations deleted due to missingness)

Multiple R-squared:  0.125575685004, Adjusted R-squared:  0.125170921396 

F-statistic: 310.244504318 on 3 and 6481 DF,  p-value: < 2.220446049e-16

[the-path-to-procrastination]

Assignment 8 (Correlation)

Again, this is an issue with the survey responses, however I choose to ignore it anyway and correlate two seemingly opposite variables (H1FS15 -You enjoyed life and H1FS16 - You felt sad).

Correlation is a rather weak -0.31, which is statistically significant p<0.001. R refuses to give a P for Spearman's even at 15 digits significance (see modified code at the end).

# Ignore all assumptions and perform correlation on categorical variables

load("~/Dropbox/coursera-PassionDrivenStatistics/addhealth_pds3.RData")

library(car)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs11 <- Recode(addhealth_pds$h1fs11, 'c(6,8 ) =  NA')

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA')

addhealth_pds$h1fs16 <- Recode(addhealth_pds$h1fs16, 'c(6,8 ) =  NA')

addhealth_pds$h1se4 <- factor(addhealth_pds$h1se4, levels = c(1,2,3,4,5,6), labels = c("moderately below average", "slightly below average", "about average", "slightly above average", "moderately above average", "extremely above average"))

# pearson rank-order correlations

library(Hmisc, pos=4)

rcorr.adjust(addhealth_pds[,c("h1fs15","h1fs16")], type="pearson")

# pearson rank-order correlations, try and get p

library(Hmisc, pos=4)

rc <- rcorr.adjust(addhealth_pds[,c("h1fs15","h1fs16")], type="pearson")

print(rc$P, digits = 20) 

[the-path-to-procrastination]

Assignment 7 (Chi Square)

Now, this was less inconvenient to run.

When examining the association between h1fs15 life enjoyment (categorical response variable) and h1se4 perception of intelligence (categorical explanatory variable), a chi-square test of independence revealed that among participants of the AddHealth study (my sample), those with self rated intelligence above average were more likely to enjoy life (84.0%) compared to those with self rated intelligence below average (75.1%), X-squared = 77.7507, df = 1, p-value < 2.2e-16

I thought of using a different R package to get a more accurate p-value (This is the smallest possible here - http://stackoverflow.com/questions/6970705/why-cant-i-get-a-p-value-smaller-than-2-2e-16) but decided against it.

.Table

               h1fs15

h1se4           less more

  below average  721 2179

  above average  573 2998

  Pearson's Chi-squared test

data:  .Table

X-squared = 77.7507, df = 1, p-value < 2.2e-16

load("~/Dropbox/coursera-PassionDrivenStatistics/addhealth_pds3.RData")

library(car)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(1,2,3 ) =  0', as.factor.result=TRUE)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(4,5,6 ) =  1', as.factor.result=TRUE)

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA')

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(0,1 ) =  0')

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(2,3 ) =  1')

addhealth_pds$h1se4 <- factor(addhealth_pds$h1se4, levels = c(0,1), labels = c("below average", "above average"))

addhealth_pds$h1fs15 <- factor(addhealth_pds$h1fs15, levels = c(0,1), labels = c("less", "more"))

.Table <- xtabs(~h1se4+h1fs15, data=addhealth_pds)

.Table

.Test <- chisq.test(.Table, correct=FALSE)

.Test

[the-path-to-procrastination]

Assignment 6

I pondered a bit on this submission. Generally, it is not a good idea to run parametric analyses on survey answers. A  number of problems do arise interpreting the results blah blah blah.

I chose to ignore that, leave my beautiful factor variables as metric and go ahead with an ANOVA that of course means very little but satisfies the course objectives. R will happily crunch the ordinal variables and demonstrate that there is a significant difference here between the groups (a difference that we did note in the pretty graph before).

When examining the association between H1FS11 -during the past week, you were happy (ordinal response variable, treated as quantitative for convenience) and H1SE4 – compared with other people your age, how intelligent are you?  (categorical explanatory variable), an Analysis of Variance (ANOVA) revealed that among young people in the Wave I of the AddHealth study (my sample), those considering themselves moderately below average intelligent reported being significantly less happy last week (Mean=1.48, s.d. 1.05) compared to all other groups, F=35.3, p<2e-16. 

Results:

              Df Sum Sq Mean Sq F value Pr(>F)    

h1se4          5    111  22.298    35.3 <2e-16 ***

Residuals   6466   4085   0.632                   

---

Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

32 observations deleted due to missingness

  > numSummary(addhealth_pds$h1fs11 , groups=addhealth_pds$h1se4, statistics=c("mean", 

+   "sd"))

                             mean        sd data:n data:NA

moderately below average 1.480519 1.0464274     77       0

slightly below average   1.859813 0.8815675    321       1

about average            2.045564 0.8045268   2502       0

slightly above average   2.164789 0.7636193   1420       0

moderately above average 2.265283 0.7656692   1734       1

extremely above average  2.232057 0.8346375    418       0

Simultaneous Tests for General Linear Hypotheses

  Multiple Comparisons of Means: Tukey Contrasts

    Fit: aov(formula = h1fs11 ~ h1se4, data = addhealth_pds)

  Linear Hypotheses:

                                                         Estimate Std. Error t value

slightly below average - moderately below average == 0    0.37929    0.10086   3.761

about average - moderately below average == 0             0.56504    0.09196   6.145

slightly above average - moderately below average == 0    0.68427    0.09300   7.358

moderately above average - moderately below average == 0  0.78476    0.09257   8.478

extremely above average - moderately below average == 0   0.75154    0.09857   7.625

about average - slightly below average == 0               0.18575    0.04712   3.942

slightly above average - slightly below average == 0      0.30498    0.04912   6.209

moderately above average - slightly below average == 0    0.40547    0.04829   8.396

extremely above average - slightly below average == 0     0.37224    0.05898   6.311

slightly above average - about average == 0               0.11923    0.02641   4.515

moderately above average - about average == 0             0.21972    0.02484   8.847

extremely above average - about average == 0              0.18649    0.04200   4.441

moderately above average - slightly above average == 0    0.10049    0.02845   3.533

extremely above average - slightly above average == 0     0.06727    0.04423   1.521

extremely above average - moderately above average == 0  -0.03323    0.04331  -0.767

                                                         Pr(>|t|)    

slightly below average - moderately below average == 0    0.00181 ** 

about average - moderately below average == 0             < 0.001 ***

slightly above average - moderately below average == 0    < 0.001 ***

moderately above average - moderately below average == 0  < 0.001 ***

extremely above average - moderately below average == 0   < 0.001 ***

about average - slightly below average == 0               < 0.001 ***

slightly above average - slightly below average == 0      < 0.001 ***

moderately above average - slightly below average == 0    < 0.001 ***

extremely above average - slightly below average == 0     < 0.001 ***

slightly above average - about average == 0               < 0.001 ***

moderately above average - about average == 0             < 0.001 ***

extremely above average - about average == 0              < 0.001 ***

moderately above average - slightly above average == 0    0.00457 ** 

extremely above average - slightly above average == 0     0.61482    

extremely above average - moderately above average == 0   0.96794    

---

Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Adjusted p values reported -- single-step method)

compact letter display:

moderately below average   slightly below average            about average 

                     "a"                      "b"                      "c" 

  slightly above average moderately above average  extremely above average 

                     "d"                      "e"                     "de" 

# Ignore all assumptions and perform ANOVA on categorical variables

load("~/Dropbox/coursera-PassionDrivenStatistics/addhealth_pds3.RData")

library(car)

addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

addhealth_pds$h1fs11 <- Recode(addhealth_pds$h1fs11, 'c(6,8 ) =  NA')

addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA')

addhealth_pds$h1fs16 <- Recode(addhealth_pds$h1fs16, 'c(6,8 ) =  NA')

addhealth_pds$h1se4 <- factor(addhealth_pds$h1se4, levels = c(1,2,3,4,5,6), labels = c("moderately below average", "slightly below average", "about average", "slightly above average", "moderately above average", "extremely above average"))

library(Rcmdr)

library(multcomp, pos=4)

library(abind, pos=4)

AnovaModel.1 <- aov(h1fs11 ~ h1se4, data=addhealth_pds)

summary(AnovaModel.1)

numSummary(addhealth_pds$h1fs11 , groups=addhealth_pds$h1se4, statistics=c("mean", "sd"))

.Pairs <- glht(AnovaModel.1, linfct = mcp(h1se4 = "Tukey"))

summary(.Pairs) # pairwise tests

confint(.Pairs) # confidence intervals

cld(.Pairs) # compact letter display

old.oma <- par(oma=c(0,5,0,0))

plot(confint(.Pairs))

par(old.oma)

remove(.Pairs)

AnovaModel.2 <- aov(h1fs15 ~ h1se4, data=addhealth_pds)

summary(AnovaModel.2)

numSummary(addhealth_pds$h1fs15 , groups=addhealth_pds$h1se4, statistics=c("mean", "sd"))

AnovaModel.3 <- aov(h1fs16 ~ h1se4, data=addhealth_pds)

summary(AnovaModel.3)

numSummary(addhealth_pds$h1fs16 , groups=addhealth_pds$h1se4, statistics=c("mean", "sd"))

[the-path-to-procrastination]

Assignment 5

This one took procrastination to a whole new level. After a lot of googling, trying out barplot options, coming up with numerous ugly and odd plots (remember: two categorical variables is never easy to plot!!!), I chose the quick and dirty way:

# just use ggplot2

> library(ggplot2)

> qplot(h1se4, data=addhealth_pds, geom="bar", fill=h1fs11, xlab = "Perceived intelligence") + labs(fill="during the past week, you were happy")

[the-path-to-procrastination]

Assignment 4

R's graphing capabilities are great; however, it takes a bit of coding to get the results you want. My code for assignment 4 required putting some rotated text labels under the X axis. This is the result:

## Graph one variable

# make size of bottom axis bigger for text labels

par(mar = c(10, 4, 4, 2) + 0.1)

# plot graph

barplot(table(addhealth_pds$h1fs11), xaxt = "n", xlab="", ylab="Frequency")

# Plot X axis labels

labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time")

text(1:4, par("usr")[3] - 0.25, srt = 45, adj = 1, labels = labels, xpd = TRUE)

# Plot X axis label

mtext(1, text = "H1FS11 during the past week, you were happy", line = 9)

[the-path-to-procrastination]

Assignment 3

Making the 96's, 98's, 6's and 8's disappear is not as simple as in SAS. There are so many packages offering a Recode option one needs to choose the prettiest. I think car does it best, as below.

> # Replace values with NA using package car

> library(car)

> addhealth_pds$h1se4 <- Recode(addhealth_pds$h1se4, 'c(96,98 ) =  NA', as.factor.result=TRUE)

> addhealth_pds$h1fs11 <- Recode(addhealth_pds$h1fs11, 'c(6,8 ) =  NA', as.factor.result=TRUE)

> addhealth_pds$h1fs15 <- Recode(addhealth_pds$h1fs15, 'c(6,8 ) =  NA', as.factor.result=TRUE)

> addhealth_pds$h1fs16 <- Recode(addhealth_pds$h1fs16, 'c(6,8 ) =  NA', as.factor.result=TRUE)

> # Get frequencies again

> summary.factor(addhealth_pds$h1se4)

   1    2    3    4    5    6 NA's 

  77  322 2502 1420 1735  418   30 

> summary.factor(addhealth_pds$h1fs11)

   0    1    2    3 NA's 

 172 1230 2690 2397   15 

> summary.factor(addhealth_pds$h1fs15)

   0    1    2    3 NA's 

 255 1043 2047 3141   18 

> summary.factor(addhealth_pds$h1fs16)

   0    1    2    3 NA's 

3405 2629  336  120   14 

[the-path-to-procrastination]

Assignment 2

My R code gains 0/1, however it looks quite pretty:

> # load data

> load("~/Dropbox/coursera-PassionDrivenStatistics/addhealth_pds3.RData")

> # Examine the structure of the data

> str(addhealth_pds)

'data.frame': 6504 obs. of  65 variables:

 $ aid    : chr  "57100270" "57101310" "57103171" "57103869" ...

 $ bio_sex: num  2 2 1 1 2 1 1 1 1 1 ...

 $ h1gi4  : num  0 0 0 0 0 0 0 0 0 0 ...

 $ h1gi6a : num  0 0 1 0 0 0 0 1 0 1 ...

 $ h1gi6b : num  1 1 0 1 1 1 1 0 1 0 ...

 $ h1gi6c : num  0 0 0 0 0 0 0 0 0 0 ...

 $ h1gi6d : num  0 0 0 0 0 0 0 0 0 0 ...

 $ h1gi20 : num  97 11 10 12 12 7 7 8 8 8 ...

 $ h1gi21 : num  4 97 97 97 97 97 97 97 97 97 ...

 $ h1ed7  : num  0 0 0 1 0 1 1 1 0 0 ...

 $ h1ed9  : num  0 0 0 0 0 0 0 0 0 0 ...

 $ h1rp1  : num  5 5 3 5 2 7 7 7 7 7 ...

 $ h1rp2  : num  1 1 3 2 2 7 7 7 7 7 ...

 $ h1rp3  : num  1 2 2 1 2 7 7 7 7 7 ...

 $ h1rp4  : num  4 4 5 1 2 7 7 7 7 7 ...

 $ h1rp5  : num  4 1 3 3 4 7 7 7 7 7 ...

 $ h1rp6  : num  5 5 5 3 4 7 7 7 7 7 ...

 $ h1se1  : num  1 1 1 1 1 97 97 97 97 97 ...

 $ h1se2  : num  1 1 1 6 1 97 97 97 97 97 ...

 $ h1se3  : num  1 5 1 4 5 97 97 97 97 97 ...

 $ h1se4  : num  6 3 3 3 3 3 5 2 5 3 ...

 $ h1fs1  : num  1 1 0 2 0 1 1 0 0 0 ...

 $ h1fs2  : num  0 1 0 1 0 0 0 1 1 0 ...

 $ h1fs3  : num  1 0 0 2 0 0 0 0 0 0 ...

 $ h1fs4  : num  3 2 3 0 1 2 1 2 1 2 ...

 $ h1fs6  : num  1 0 0 1 0 0 0 0 0 1 ...

 $ h1fs7  : num  1 1 1 1 0 2 1 1 1 0 ...

 $ h1fs8  : num  3 3 3 2 2 3 3 1 2 1 ...

 $ h1fs9  : num  1 0 0 0 0 1 0 0 1 0 ...

 $ h1fs10 : num  0 1 0 1 0 0 0 0 0 0 ...

 $ h1fs11 : num  3 3 3 2 3 0 1 2 1 2 ...

 $ h1fs12 : num  1 0 0 1 2 1 0 0 0 0 ...

 $ h1fs13 : num  1 0 0 0 1 0 0 0 0 0 ...

 $ h1fs14 : num  0 1 0 1 0 0 1 0 0 0 ...

 $ h1fs15 : num  3 3 3 1 2 2 0 3 3 3 ...

 $ h1fs16 : num  1 0 0 0 1 0 0 0 1 1 ...

 $ h1fs17 : num  0 1 0 1 0 1 0 1 0 0 ...

 $ h1fs18 : num  0 1 0 1 0 0 1 1 0 0 ...

 $ h1fs19 : num  1 0 0 0 0 1 0 0 0 0 ...

 $ h1rr1  : num  1 1 0 1 1 0 1 1 1 0 ...

 $ h1nr1  : num  0 0 1 1 0 0 0 1 1 1 ...

 $ h1nr2  : num  1 1 0 0 1 0 0 1 0 0 ...

 $ h1nr3  : num  0 0 0 0 0 0 0 0 0 0 ...

 $ h1nr5  : num  1 1 0 1 1 0 0 0 1 0 ...

 $ h1nr6  : num  4 4 997 3 3 997 997 997 5 997 ...

 $ h1jo10 : num  1 0 7 7 0 7 7 0 0 7 ...

 $ h1fp7  : num  1 1 7 7 1 7 7 7 7 7 ...

 $ h1fp8  : num  1 1 97 97 1 97 97 97 97 97 ...

 $ h1fp9m : num  98 5 97 97 8 97 97 97 97 97 ...

 $ h1su1  : num  0 0 0 1 1 0 0 0 0 0 ...

 $ h1su2  : num  7 7 7 0 0 7 7 7 7 7 ...

 $ h1pa1  : num  1 4 1 1 4 1 1 1 1 1 ...

 $ h1pa2  : num  2 4 1 5 4 1 1 2 1 1 ...

 $ h1pa3  : num  3 4 5 1 4 1 1 4 1 1 ...

 $ h1pa4  : num  3 4 1 7 7 1 1 7 7 7 ...

 $ h1pa5  : num  3 4 1 7 7 1 1 7 7 7 ...

 $ h1pa6  : num  3 4 5 7 7 1 3 7 7 7 ...

 $ h1pa7  : num  0 1 0 1 0 0 0 0 1 1 ...

 $ h1re1  : num  28 0 4 0 5 5 4 22 2 19 ...

 $ h1re2  : num  1 7 1 7 1 1 1 1 1 1 ...

 $ h1re3  : num  1 7 3 7 4 1 3 3 3 2 ...

 $ h1re4  : num  1 7 1 7 1 1 1 2 3 2 ...

 $ h1re5  : num  7 7 8 7 1 1 8 7 0 0 ...

 $ h1re6  : num  2 7 3 7 4 1 4 1 3 3 ...

 $ h1re7  : num  1 7 3 7 4 1 2 4 2 4 ...

 - attr(*, "datalabel")= chr ""

 - attr(*, "time.stamp")= chr ""

 - attr(*, "formats")= chr  "%8s" "%12.0g" "%12.0g" "%12.0g" ...

 - attr(*, "types")= int  8 255 255 255 255 255 255 255 255 255 ...

 - attr(*, "val.labels")= chr  "" "" "" "" ...

 - attr(*, "var.labels")= chr  "RESPONDENT IDENTIFIER" "BIOLOGICAL SEX-W1" "S1Q4 ARE YOU OF HISPANIC ORIGIN-W1" "S1Q6A RACE-WHITE-W1" ...

 - attr(*, "version")= int -7

> summary(addhealth_pds)

     aid               bio_sex          h1gi4            h1gi6a      

 Length:6504        Min.   :1.000   Min.   :0.0000   Min.   :0.0000  

 Class :character   1st Qu.:1.000   1st Qu.:0.0000   1st Qu.:0.0000  

 Mode  :character   Median :2.000   Median :0.0000   Median :1.0000  

                    Mean   :1.517   Mean   :0.1416   Mean   :0.6823  

                    3rd Qu.:2.000   3rd Qu.:0.0000   3rd Qu.:1.0000  

                    Max.   :6.000   Max.   :8.0000   Max.   :8.0000  

                                                                           h1gi6b           h1gi6c            h1gi6d            h1gi20    

 Min.   :0.0000   Min.   :0.00000   Min.   :0.00000   Min.   : 7.0  

 1st Qu.:0.0000   1st Qu.:0.00000   1st Qu.:0.00000   1st Qu.: 8.0  

 Median :0.0000   Median :0.00000   Median :0.00000   Median :10.0  

 Mean   :0.2711   Mean   :0.05843   Mean   :0.06365   Mean   :11.8  

 3rd Qu.:1.0000   3rd Qu.:0.00000   3rd Qu.:0.00000   3rd Qu.:11.0  

 Max.   :8.0000   Max.   :8.00000   Max.   :8.00000   Max.   :99.0  

                                                                          h1gi21         h1ed7            h1ed9             h1rp1      

 Min.   : 1.0   Min.   :0.0000   Min.   :0.00000   Min.   :1.000  

 1st Qu.:97.0   1st Qu.:0.0000   1st Qu.:0.00000   1st Qu.:1.000  

 Median :97.0   Median :0.0000   Median :0.00000   Median :2.000  

 Mean   :95.2   Mean   :0.2915   Mean   :0.05711   Mean   :3.373  

 3rd Qu.:97.0   3rd Qu.:1.0000   3rd Qu.:0.00000   3rd Qu.:7.000  

 Max.   :98.0   Max.   :8.0000   Max.   :8.00000   Max.   :9.000  

                NA's   :3        NA's   :8                        

     h1rp2           h1rp3           h1rp4           h1rp5      

 Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   :1.000  

 1st Qu.:4.000   1st Qu.:1.000   1st Qu.:4.000   1st Qu.:3.000  

 Median :5.000   Median :2.000   Median :5.000   Median :4.000  

 Mean   :5.096   Mean   :3.308   Mean   :4.914   Mean   :4.441  

 3rd Qu.:7.000   3rd Qu.:7.000   3rd Qu.:7.000   3rd Qu.:7.000  

 Max.   :9.000   Max.   :8.000   Max.   :9.000   Max.   :9.000  

                                                                      h1rp6           h1se1          h1se2           h1se3      

 Min.   :1.000   Min.   : 1.0   Min.   : 1.00   Min.   : 1.00  

 1st Qu.:3.000   1st Qu.: 1.0   1st Qu.: 1.00   1st Qu.: 1.00  

 Median :4.000   Median : 3.0   Median : 2.00   Median : 2.00  

 Mean   :4.638   Mean   :33.6   Mean   :32.96   Mean   :33.09  

 3rd Qu.:7.000   3rd Qu.:97.0   3rd Qu.:97.00   3rd Qu.:97.00  

 Max.   :9.000   Max.   :99.0   Max.   :99.00   Max.   :99.00  

                                                                     h1se4            h1fs1            h1fs2            h1fs3       

 Min.   : 1.000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  

 1st Qu.: 3.000   1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:0.0000  

 Median : 4.000   Median :0.0000   Median :0.0000   Median :0.0000  

 Mean   : 4.308   Mean   :0.5151   Mean   :0.4783   Mean   :0.4077  

 3rd Qu.: 5.000   3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:1.0000  

 Max.   :98.000   Max.   :8.0000   Max.   :8.0000   Max.   :8.0000  

                                                                          h1fs4          h1fs6            h1fs7            h1fs8      

 Min.   :0.00   Min.   :0.0000   Min.   :0.0000   Min.   :0.000  

 1st Qu.:1.00   1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:1.000  

 Median :2.00   Median :0.0000   Median :1.0000   Median :2.000  

 Mean   :1.95   Mean   :0.5326   Mean   :0.7414   Mean   :1.871  

 3rd Qu.:3.00   3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:3.000  

 Max.   :8.00   Max.   :8.0000   Max.   :8.0000   Max.   :8.000  

                                                                       h1fs9           h1fs10           h1fs11          h1fs12     

 Min.   :0.000   Min.   :0.0000   Min.   :0.000   Min.   :0.000  

 1st Qu.:0.000   1st Qu.:0.0000   1st Qu.:2.000   1st Qu.:0.000  

 Median :0.000   Median :0.0000   Median :2.000   Median :0.000  

 Mean   :0.238   Mean   :0.3364   Mean   :2.139   Mean   :0.581  

 3rd Qu.:0.000   3rd Qu.:1.0000   3rd Qu.:3.000   3rd Qu.:1.000  

 Max.   :8.000   Max.   :8.0000   Max.   :8.000   Max.   :8.000  

                                                                       h1fs13           h1fs14           h1fs15          h1fs16      

 Min.   :0.0000   Min.   :0.0000   Min.   :0.000   Min.   :0.0000  

 1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:2.000   1st Qu.:0.0000  

 Median :0.0000   Median :0.0000   Median :2.000   Median :0.0000  

 Mean   :0.4839   Mean   :0.4182   Mean   :2.258   Mean   :0.5783  

 3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:3.000   3rd Qu.:1.0000  

 Max.   :8.0000   Max.   :8.0000   Max.   :8.000   Max.   :8.0000  

                                                                         h1fs17           h1fs18           h1fs19           h1rr1       

 Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  

 1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:0.0000  

 Median :0.0000   Median :1.0000   Median :0.0000   Median :1.0000  

 Mean   :0.4394   Mean   :0.6359   Mean   :0.1799   Mean   :0.5984  

 3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:0.0000   3rd Qu.:1.0000  

 Max.   :8.0000   Max.   :8.0000   Max.   :8.0000   Max.   :9.0000  

                                                                          h1nr1            h1nr2            h1nr3             h1nr5       

 Min.   :0.0000   Min.   :0.0000   Min.   :0.00000   Min.   :0.0000  

 1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:0.00000   1st Qu.:0.0000  

 Median :0.0000   Median :0.0000   Median :0.00000   Median :0.0000  

 Mean   :0.5197   Mean   :0.5466   Mean   :0.07411   Mean   :0.3538  

 3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:0.00000   3rd Qu.:1.0000  

 Max.   :9.0000   Max.   :9.0000   Max.   :9.00000   Max.   :9.0000  

                                                                           h1nr6           h1jo10          h1fp7          h1fp8      

 Min.   :  1.0   Min.   :0.000   Min.   :0.00   Min.   : 1.00  

 1st Qu.:153.8   1st Qu.:0.000   1st Qu.:7.00   1st Qu.:97.00  

 Median :997.0   Median :1.000   Median :7.00   Median :97.00  

 Mean   :748.8   Mean   :3.238   Mean   :5.69   Mean   :93.57  

 3rd Qu.:997.0   3rd Qu.:7.000   3rd Qu.:7.00   3rd Qu.:97.00  

 Max.   :999.0   Max.   :8.000   Max.   :8.00   Max.   :98.00  

                                                                     h1fp9m          h1su1            h1su2          h1pa1      

 Min.   : 1.00   Min.   :0.0000   Min.   :0.00   Min.   :1.000  

 1st Qu.:97.00   1st Qu.:0.0000   1st Qu.:7.00   1st Qu.:1.000  

 Median :97.00   Median :0.0000   Median :7.00   Median :1.000  

 Mean   :93.85   Mean   :0.1971   Mean   :6.18   Mean   :2.057  

 3rd Qu.:97.00   3rd Qu.:0.0000   3rd Qu.:7.00   3rd Qu.:3.000  

 Max.   :98.00   Max.   :9.0000   Max.   :8.00   Max.   :9.000  

                                                                      h1pa2          h1pa3           h1pa4           h1pa5      

 Min.   :1.00   Min.   :1.000   Min.   :1.000   Min.   :1.000  

 1st Qu.:1.00   1st Qu.:1.000   1st Qu.:1.000   1st Qu.:1.000  

 Median :2.00   Median :3.000   Median :2.000   Median :3.000  

 Mean   :2.34   Mean   :3.153   Mean   :3.493   Mean   :3.603  

 3rd Qu.:3.00   3rd Qu.:5.000   3rd Qu.:7.000   3rd Qu.:7.000  

 Max.   :9.00   Max.   :9.000   Max.   :9.000   Max.   :9.000  

                                                                     h1pa6           h1pa7            h1re1          h1re2      

 Min.   :1.000   Min.   :0.0000   Min.   : 0.0   Min.   :1.000  

 1st Qu.:2.000   1st Qu.:0.0000   1st Qu.: 4.0   1st Qu.:1.000  

 Median :4.000   Median :0.0000   Median :10.0   Median :1.000  

 Mean   :4.129   Mean   :0.3493   Mean   :13.1   Mean   :2.292  

 3rd Qu.:7.000   3rd Qu.:0.0000   3rd Qu.:22.0   3rd Qu.:2.000  

 Max.   :9.000   Max.   :9.0000   Max.   :99.0   Max.   :8.000  

                                                                      h1re3           h1re4           h1re5           h1re6      

 Min.   :1.000   Min.   :1.000   Min.   :0.000   Min.   :1.000  

 1st Qu.:1.000   1st Qu.:1.000   1st Qu.:0.000   1st Qu.:1.000  

 Median :2.000   Median :2.000   Median :1.000   Median :2.000  

 Mean   :2.667   Mean   :2.372   Mean   :3.305   Mean   :2.711  

 3rd Qu.:3.000   3rd Qu.:2.000   3rd Qu.:7.000   3rd Qu.:4.000  

 Max.   :8.000   Max.   :8.000   Max.   :9.000   Max.   :8.000  

                                                                      h1re7      

 Min.   :1.000  

 1st Qu.:2.000  

 Median :4.000  

 Mean   :3.347  

 3rd Qu.:4.000  

 Max.   :8.000  

                 > describe(addhealth_pds)

addhealth_pds 

 65  Variables      6504  Observations

------------------------------------------------------------------------

aid 

      n missing  unique 

   6504       0    6504 

lowest : 57100270 57101310 57103171 57103869 57104553

highest: 99719930 99719939 99719970 99719976 99719978 

------------------------------------------------------------------------

bio_sex 

      n missing  unique    Mean 

   6504       0       3   1.517 

1 (3147, 48%), 2 (3356, 52%), 6 (1, 0%) 

------------------------------------------------------------------------

h1gi4 

      n missing  unique    Mean 

   6504       0       4  0.1416 

0 (5738, 88%), 1 (743, 11%), 6 (3, 0%), 8 (20, 0%) 

------------------------------------------------------------------------

h1gi6a 

      n missing  unique    Mean 

   6504       0       4  0.6823 

0 (2191, 34%), 1 (4294, 66%), 6 (4, 0%), 8 (15, 0%) 

------------------------------------------------------------------------

h1gi6b 

      n missing  unique    Mean 

   6504       0       4  0.2711 

0 (4866, 75%), 1 (1619, 25%), 6 (4, 0%), 8 (15, 0%) 

------------------------------------------------------------------------

h1gi6c 

      n missing  unique    Mean 

   6504       0       4 0.05843 

0 (6249, 96%), 1 (236, 4%), 6 (4, 0%), 8 (15, 0%) 

------------------------------------------------------------------------

h1gi6d 

      n missing  unique    Mean 

   6504       0       4 0.06365 

0 (6215, 96%), 1 (270, 4%), 6 (4, 0%), 8 (15, 0%) 

------------------------------------------------------------------------

h1gi20 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      10    11.8       7       7       8      10      11 

    .90     .95 

     12      12 

            7   8    9   10   11  12 96  97 98 99

Frequency 979 992 1107 1144 1122 993  1 128  3 35

%          15  15   17   18   17  15  0   2  0  1

------------------------------------------------------------------------

h1gi21 

      n missing  unique    Mean 

   6504       0       9    95.2 

           1  2 3  4 5  6 96   97 98

Frequency 11 42 7 33 5 27  1 6376  2

%          0  1 0  1 0  0  0   98  0

------------------------------------------------------------------------

h1ed7 

      n missing  unique    Mean 

   6501       3       4  0.2915 

0 (4687, 72%), 1 (1801, 28%), 6 (5, 0%), 8 (8, 0%) 

------------------------------------------------------------------------

h1ed9 

      n missing  unique    Mean 

   6496       8       4 0.05711 

0 (6187, 95%), 1 (299, 5%), 6 (4, 0%), 8 (6, 0%) 

------------------------------------------------------------------------

h1rp1 

      n missing  unique    Mean 

   6504       0       9   3.373 

             1    2   3   4  5 6    7  8 9

Frequency 2615 1112 356 253 91 8 2039 29 1

%           40   17   5   4  1 0   31  0 0

------------------------------------------------------------------------

h1rp2 

      n missing  unique    Mean 

   6504       0       9   5.096 

           1   2   3    4    5 6    7  8 9

Frequency 94 312 403 1435 2183 7 2039 30 1

%          1   5   6   22   34 0   31  0 0

------------------------------------------------------------------------

h1rp3 

      n missing  unique    Mean 

   6504       0       8   3.308 

             1    2   3   4  5 6    7  8

Frequency 2788 1065 339 157 78 6 2039 32

%           43   16   5   2  1 0   31  0

------------------------------------------------------------------------

h1rp4 

      n missing  unique    Mean 

   6504       0       9   4.914 

            1   2   3    4    5 6    7  8 9

Frequency 346 464 321 1392 1882 5 2039 54 1

%           5   7   5   21   29 0   31  1 0

------------------------------------------------------------------------

h1rp5 

      n missing  unique    Mean 

   6504       0       9   4.441 

            1   2    3    4   5 6    7  8 9

Frequency 294 553 1930 1218 398 8 2039 63 1

%           5   9   30   19   6 0   31  1 0

------------------------------------------------------------------------

h1rp6 

      n missing  unique    Mean 

   6504       0       9   4.638 

            1   2    3    4   5  6    7  8 9

Frequency 314 564 1069 1552 891 10 2039 63 2

%           5   9   16   24  14  0   31  1 0

------------------------------------------------------------------------

h1se1 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      10    33.6       1       1       1       3      97 

    .90     .95 

     97      97 

             1   2   3   4   5   6 96   97  98 99

Frequency 2159 881 758 201 213 131 20 2039 101  1

%           33  14  12   3   3   2  0   31   2  0

------------------------------------------------------------------------

h1se2 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      10   32.96       1       1       1       2      97 

    .90     .95 

     97      97 

             1   2   3  4   5  6 96   97 98 99

Frequency 2722 934 412 88 115 99 20 2039 74  1

%           42  14   6  1   2  2  0   31  1  0

------------------------------------------------------------------------

h1se3 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      10   33.09       1       1       1       2      97 

    .90     .95 

     97      97 

             1   2   3   4   5  6 96   97 98 99

Frequency 2521 755 617 195 226 56 15 2039 79  1

%           39  12   9   3   3  1  0   31  1  0

------------------------------------------------------------------------

h1se4 

      n missing  unique    Mean 

   6504       0       8   4.308 

           1   2    3    4    5   6 96 98

Frequency 77 322 2502 1420 1735 418  5 25

%          1   5   38   22   27   6  0  0

------------------------------------------------------------------------

h1fs1 

      n missing  unique    Mean 

   6504       0       6  0.5151 

             0    1   2   3 6  8

Frequency 3913 2068 385 116 6 16

%           60   32   6   2 0  0

------------------------------------------------------------------------

h1fs2 

      n missing  unique    Mean 

   6504       0       6  0.4783 

             0    1   2   3 6  8

Frequency 4192 1744 410 141 6 11

%           64   27   6   2 0  0

------------------------------------------------------------------------

h1fs3 

      n missing  unique    Mean 

   6504       0       6  0.4077 

             0    1   2   3 6  8

Frequency 4668 1296 372 144 6 18

%           72   20   6   2 0  0

------------------------------------------------------------------------

h1fs4 

      n missing  unique    Mean 

   6504       0       6    1.95 

            0    1    2    3 6  8

Frequency 715 1353 2070 2345 6 15

%          11   21   32   36 0  0

------------------------------------------------------------------------

h1fs6 

      n missing  unique    Mean 

   6504       0       6  0.5326 

             0    1   2   3 6  8

Frequency 3994 1853 444 193 8 12

%           61   28   7   3 0  0

------------------------------------------------------------------------

h1fs7 

      n missing  unique    Mean 

   6504       0       6  0.7414 

             0    1   2   3 6  8

Frequency 2755 2934 630 168 6 11

%           42   45  10   3 0  0

------------------------------------------------------------------------

h1fs8 

      n missing  unique    Mean 

   6504       0       6   1.871 

            0    1    2    3 6  8

Frequency 720 1567 2185 2003 6 23

%          11   24   34   31 0  0

------------------------------------------------------------------------

h1fs9 

      n missing  unique    Mean 

   6504       0       6   0.238 

             0   1   2  3 6  8

Frequency 5451 782 164 80 9 18

%           84  12   3  1 0  0

------------------------------------------------------------------------

h1fs10 

      n missing  unique    Mean 

   6504       0       6  0.3364 

             0    1   2  3 6  8

Frequency 4714 1545 163 65 7 10

%           72   24   3  1 0  0

------------------------------------------------------------------------

h1fs11 

      n missing  unique    Mean 

   6504       0       6   2.139 

            0    1    2    3 6 8

Frequency 172 1230 2690 2397 6 9

%           3   19   41   37 0 0

------------------------------------------------------------------------

h1fs12 

      n missing  unique    Mean 

   6504       0       6   0.581 

             0    1   2   3 6  8

Frequency 3642 2206 476 161 7 12

%           56   34   7   2 0  0

------------------------------------------------------------------------

h1fs13 

      n missing  unique    Mean 

   6504       0       6  0.4839 

             0    1   2   3 6  8

Frequency 4157 1787 401 140 7 12

%           64   27   6   2 0  0

------------------------------------------------------------------------

h1fs14 

      n missing  unique    Mean 

   6504       0       6  0.4182 

             0    1   2  3 6 8

Frequency 4307 1839 256 87 6 9

%           66   28   4  1 0 0

------------------------------------------------------------------------

h1fs15 

      n missing  unique    Mean 

   6504       0       6   2.258 

            0    1    2    3 6  8

Frequency 255 1043 2047 3141 8 10

%           4   16   31   48 0  0

------------------------------------------------------------------------

h1fs16 

      n missing  unique    Mean 

   6504       0       6  0.5783 

             0    1   2   3 6 8

Frequency 3405 2629 336 120 6 8

%           52   40   5   2 0 0

------------------------------------------------------------------------

h1fs17 

      n missing  unique    Mean 

   6504       0       6  0.4394 

             0    1   2   3 6  8

Frequency 4246 1859 276 105 6 12

%           65   29   4   2 0  0

------------------------------------------------------------------------

h1fs18 

      n missing  unique    Mean 

   6504       0       6  0.6359 

             0    1   2  3 6  8

Frequency 3124 2814 462 84 7 13

%           48   43   7  1 0  0

------------------------------------------------------------------------

h1fs19 

      n missing  unique    Mean 

   6504       0       6  0.1799 

             0   1   2  3 6  8

Frequency 5728 545 149 63 7 12

%           88   8   2  1 0  0

------------------------------------------------------------------------

h1rr1 

      n missing  unique    Mean 

   6504       0       5  0.5984 

             0    1  6  8 9

Frequency 2878 3582 22 20 2

%           44   55  0  0 0

------------------------------------------------------------------------

h1nr1 

      n missing  unique    Mean 

   6504       0       5  0.5197 

             0    1  6  8 9

Frequency 3606 2817 43 37 1

%           55   43  1  1 0

------------------------------------------------------------------------

h1nr2 

      n missing  unique    Mean 

   6504       0       5  0.5466 

             0    1  6  8 9

Frequency 3283 3164 33 23 1

%           50   49  1  0 0

------------------------------------------------------------------------

h1nr3 

      n missing  unique    Mean 

   6504       0       5 0.07411 

             0  1  6  8 9

Frequency 6367 79 31 26 1

%           98  1  0  0 0

------------------------------------------------------------------------

h1nr5 

      n missing  unique    Mean 

   6504       0       5  0.3538 

             0    1  6  8 9

Frequency 4726 1687 58 31 2

%           73   26  1  0 0

------------------------------------------------------------------------

h1nr6 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      50   748.8     2.0     3.0   153.8   997.0   997.0 

    .90     .95 

  997.0   997.0 

lowest :   1   2   3   4   5, highest: 500 996 997 998 999 

------------------------------------------------------------------------

h1jo10 

      n missing  unique    Mean 

   6504       0       5   3.238 

             0   1 6    7 8

Frequency 3132 423 4 2944 1

%           48   7 0   45 0

------------------------------------------------------------------------

h1fp7 

      n missing  unique    Mean 

   6504       0       5    5.69 

             0   1 6    7 8

Frequency 1014 236 6 5245 3

%           16   4 0   81 0

------------------------------------------------------------------------

h1fp8 

      n missing  unique    Mean 

   6504       0       6   93.57 

            1  2 3 4   97 98

Frequency 191 33 6 3 6268  3

%           3  1 0 0   96  0

------------------------------------------------------------------------

h1fp9m 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      15   93.85      97      97      97      97      97 

    .90     .95 

     97      97 

           1  2  3  4  5  6  7  8 9 10 11 12 96   97 98

Frequency 19 13 24 22 24 21 11 27 7 18 21 19  1 6268  9

%          0  0  0  0  0  0  0  0 0  0  0  0  0   96  0

------------------------------------------------------------------------

h1su1 

      n missing  unique    Mean 

   6504       0       5  0.1971 

             0   1  6  8 9

Frequency 5614 821 46 22 1

%           86  13  1  0 0

------------------------------------------------------------------------

h1su2 

      n missing  unique    Mean 

   6504       0       8    6.18 

            0   1  2  3  4 6    7 8

Frequency 587 131 64 10 25 3 5683 1

%           9   2  1  0  0 0   87 0

------------------------------------------------------------------------

h1pa1 

      n missing  unique    Mean 

   6504       0       9   2.057 

             1    2   3   4  5  6   7  8 9

Frequency 3359 1514 971 164 38 26 381 50 1

%           52   23  15   3  1  0   6  1 0

------------------------------------------------------------------------

h1pa2 

      n missing  unique    Mean 

   6504       0       9    2.34 

             1    2    3   4  5  6   7  8 9

Frequency 2642 1560 1253 502 84 27 381 54 1

%           41   24   19   8  1  0   6  1 0

------------------------------------------------------------------------

h1pa3 

      n missing  unique    Mean 

   6504       0       9   3.153 

             1   2    3    4    5  6   7  8 9

Frequency 1805 881 1003 1148 1186 29 381 70 1

%           28  14   15   18   18  0   6  1 0

------------------------------------------------------------------------

h1pa4 

      n missing  unique    Mean 

   6504       0       9   3.493 

             1   2   3   4  5  6    7  8 9

Frequency 2583 876 623 137 26 20 2198 39 2

%           40  13  10   2  0  0   34  1 0

------------------------------------------------------------------------

h1pa5 

      n missing  unique    Mean 

   6504       0       9   3.603 

             1   2   3   4  5  6    7  8 9

Frequency 2263 952 730 255 48 19 2198 37 2

%           35  15  11   4  1  0   34  1 0

------------------------------------------------------------------------

h1pa6 

      n missing  unique    Mean 

   6504       0       9   4.129 

             1   2   3   4   5  6    7  8 9

Frequency 1597 569 764 620 683 19 2198 52 2

%           25   9  12  10  11  0   34  1 0

------------------------------------------------------------------------

h1pa7 

      n missing  unique    Mean 

   6504       0       6  0.3493 

             0    1  6  7  8 9

Frequency 4957 1432 24 33 57 1

%           76   22  0  1  1 0

------------------------------------------------------------------------

h1re1 

      n missing  unique    Mean     .05     .10     .25     .50     .75 

   6504       0      32    13.1       0       0       4      10      22 

    .90     .95 

     22      27 

lowest :  0  1  2  3  4, highest: 27 28 96 98 99 

------------------------------------------------------------------------

h1re2 

      n missing  unique    Mean 

   6504       0       6   2.292 

             1    2   3  6   7   8

Frequency 4138 1105 119 26 879 237

%           64   17   2  0  14   4

------------------------------------------------------------------------

h1re3 

      n missing  unique    Mean 

   6504       0       7   2.667 

             1    2    3   4 6   7 8

Frequency 2567 1274 1105 664 8 879 7

%           39   20   17  10 0  14 0

------------------------------------------------------------------------

h1re4 

      n missing  unique    Mean 

   6504       0       7   2.372 

             1    2   3   4 6   7 8

Frequency 2812 2218 391 193 3 879 8

%           43   34   6   3 0  14 0

------------------------------------------------------------------------

h1re5 

      n missing  unique    Mean 

   6504       0       6   3.305 

             0    1  6    7   8 9

Frequency 1918 1784 11 2681 109 1

%           29   27  0   41   2 0

------------------------------------------------------------------------

h1re6 

      n missing  unique    Mean 

   6504       0       8   2.711 

             1    2   3   4   5 6   7 8

Frequency 2722 1370 586 496 440 6 879 5

%           42   21   9   8   7 0  14 0

------------------------------------------------------------------------

h1re7 

      n missing  unique    Mean 

   6504       0       7   3.347 

             1   2   3    4 6   7 8

Frequency 1403 926 862 2421 6 879 7

%           22  14  13   37 0  14 0

------------------------------------------------------------------------

> # Attach variable labels

> var.labels <- attr(addhealth_pds,"var.labels")

> data.key <- data.frame(var.name=names(addhealth_pds),var.labels)

> data.key

   var.name                               var.labels

1       aid                    RESPONDENT IDENTIFIER

2   bio_sex                        BIOLOGICAL SEX-W1

3     h1gi4       S1Q4 ARE YOU OF HISPANIC ORIGIN-W1

4    h1gi6a                      S1Q6A RACE-WHITE-W1

5    h1gi6b           S1Q6B RACE-AFRICAN AMERICAN-W1

6    h1gi6c            S1Q6C RACE-AMERICAN INDIAN-W1

7    h1gi6d                      S1Q6D RACE-ASIAN-W1

8    h1gi20                           S1Q20 GRADE-W1

9    h1gi21               S1Q21 WHY NOT IN SCHOOL-W1

10    h1ed7  S5Q7 RECEIVED OUT-OF-SCHL SUSPENSION-W1

11    h1ed9             S5Q9 EXPELLED FROM SCHOOL-W1

12    h1rp1    S8Q1 PREGNANT NOW ONE OF THE WORST-W1

13    h1rp2          S8Q2 PREGNANT NOW NOT SO BAD-W1

14    h1rp3      S8Q3 WILL SUFFER IF HIV POSITIVE-W1

15    h1rp4          S8Q4 PROTECTION A BIG HASSLE-W1

16    h1rp5 S8Q5 RISK OF PREGNANCY W/O PROTECTION-W1

17    h1rp6      S8Q6 RISK OF AIDS W/O PROTECTION-W1

18    h1se1     S9Q1 USE BIRTH CONTROL IF AROUSED-W1

19    h1se2   S9Q2 BIRTHCONTROL-ADVANCED PLANNING-W1

20    h1se3   S9Q3 RESIST SEX IF NO BIRTH CONTROL-W1

21    h1se4                S9Q4 YOUR INTELLIGENCE-W1

22    h1fs1              S10Q1 BOTHERED BY THINGS-W1

23    h1fs2                   S10Q2 POOR APPETITE-W1

24    h1fs3                   S10Q3 HAD THE BLUES-W1

25    h1fs4    S10Q4 JUST AS GOOD AS OTHER PEOPLE-W1

26    h1fs6                  S10Q6 FELT DEPRESSED-W1

27    h1fs7          S10Q7 TOO TIRED TO DO THINGS-W1

28    h1fs8        S10Q8 HOPEFUL ABOUT THE FUTURE-W1

29    h1fs9         S10Q9 LIFE HAD BEEN A FAILURE-W1

30   h1fs10                        S10Q10 FEARFUL-W1

31   h1fs11                          S10Q11 HAPPY-W1

32   h1fs12         S10Q12 TALKED LESS THAN USUAL-W1

33   h1fs13                    S10Q13 FELT LONELY-W1

34   h1fs14       S10Q14 PEOPLE UNFRIENDLY TO YOU-W1

35   h1fs15                   S10Q15 ENJOYED LIFE-W1

36   h1fs16                       S10Q16 FELT SAD-W1

37   h1fs17        S10Q17 FELT PEOPLE DISLIKE YOU-W1

38   h1fs18     S10Q18 HARD TO START DOING THINGS-W1

39   h1fs19          S10Q19 LIFE NOT WORTH LIVING-W1

40    h1rr1          S22Q1 DID YOU HAVE A ROMANCE-W1

41    h1nr1      S26Q1 EVER ATTRACTED TO A FEMALE-W1

42    h1nr2        S26Q2 EVER ATTRACTED TO A MALE-W1

43    h1nr3          S26Q3 SEX FOR DRUGS OR MONEY-W1

44    h1nr5       S26Q5 NON-ROMANCE SEX W/ ANYONE-W1

45    h1nr6        S26Q6 TOTAL NO OF SEX PARTNERS-W1

46   h1jo10                S30Q10 DRUNK AT SCHOOL-W1

47    h1fp7             S32BQ7 EVER BEEN PREGNANT-W1

48    h1fp8          S32BQ8 NUMBER OF PREGNANCIES-W1

49   h1fp9m        S32BQ9M MOST RECENT PREG-MONTH-W1

50    h1su1     S33Q1 PAST YR-THINK ABOUT SUICIDE-W1

51    h1su2         S33Q2 PAST YR-ATTEMPT SUICIDE-W1

52    h1pa1     S34Q1 MOM-FEEL ABOUT YOUR SEXLIFE-W1

53    h1pa2  S34Q2 MOM-YOU HAVING SEX WITH STEADY-W1

54    h1pa3         S34Q3 MOM-USE OF BIRTHCONTROL-W1

55    h1pa4     S34Q4 DAD-FEEL ABOUT YOUR SEXLIFE-W1

56    h1pa5  S34Q5 DAD-YOU HAVING SEX WITH STEADY-W1

57    h1pa6         S34Q6 DAD-USE OF BIRTHCONTROL-W1

58    h1pa7 S34Q7 CONSIDER HAVING CHILD UNMARRIED-W1

59    h1re1                        S37Q1 RELIGION-W1

60    h1re2  S37Q2 SCRIPTURES ARE THE WORD OF GOD-W1

61    h1re3         S37Q3 PAST YR-ATTEND SERVICES-W1

62    h1re4       S37Q4 HOW IMPORTANT IS RELIGION-W1

63    h1re5            S37Q5 BORN-AGAIN CHRISTIAN-W1

64    h1re6           S37Q6 HOW OFTEN DO YOU PRAY-W1

65    h1re7     S37Q7 PAST YR-ATTEND YOUTH GROUPS-W1

> # Make nominal variables factors

> h1se4 <- factor(addhealth_pds$h1se4)

> h1fs11 <- factor(addhealth_pds$h1fs11)

> h1fs15 <- factor(addhealth_pds$h1fs15)

> h1fs16 <- factor(addhealth_pds$h1fs16)

> # Attach value labels

> val.labels <- attr(addhealth_pds,"val.labels")

> addhealth_pds$h1se4 <- factor(addhealth_pds$h1se4, levels = c(1,2,3,4,5,6,96,98), labels = c("moderately below average", "slightly below average", "about average", "slightly above average", "moderately above average", "extremely above average","refused", "don’t know"))

> addhealth_pds$h1fs11 <- factor(addhealth_pds$h1fs11, levels = c(0,1,2,3,6,8), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time", "refused", "don’t know"))

> addhealth_pds$h1fs15 <- factor(addhealth_pds$h1fs15, levels = c(0,1,2,3,6,8), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time", "refused", "don’t know"))

> addhealth_pds$h1fs16 <- factor(addhealth_pds$h1fs16, levels = c(0,1,2,3,6,8), labels = c("never or rarely", "sometimes", "a lot of the time", "most of the time or all of the time", "refused", "don’t know"))

> # Get frequencies

> summary.factor(addhealth_pds$h1se4)

moderately below average   slightly below average 

                      77                      322 

           about average   slightly above average 

                    2502                     1420 

moderately above average  extremely above average 

                    1735                      418 

                 refused               don’t know 

                       5                       25 

> summary.factor(addhealth_pds$h1fs11)

                    never or rarely                           sometimes 

                                172                                1230 

                  a lot of the time most of the time or all of the time 

                               2690                                2397 

                            refused                          don’t know 

                                  6                                   9 

> summary.factor(addhealth_pds$h1fs15)

                    never or rarely                           sometimes 

                                255                                1043 

                  a lot of the time most of the time or all of the time 

                               2047                                3141 

                            refused                          don’t know 

                                  8                                  10 

> summary.factor(addhealth_pds$h1fs16)

                    never or rarely                           sometimes 

                               3405                                2629 

                  a lot of the time most of the time or all of the time 

                                336                                 120 

                            refused                          don’t know 

                                  6                                   8 

[the-path-to-procrastination]

Let's work on the assignments - with R code

Amazing! R has been upgraded to version 3.0.0 (2013-04-03) -- "Masked Marvel" since I set out on this project.

The analysis will be done partly on Mac OS X (Mountain Lion) on a MacBook Air sporting a 1.8GHz Core i5 with 8GB RAM. Partly on a Windows 7 virtual machine, partly on a Ubuntu virtual machine. If I get fed up with R I might go back to SPSS or Stata halfway through - bear with me.

[1] NOTE: R Commander Version 1.9-6: Wed May 8 16:48:20 2013

[2] NOTE: The dataset addhealth_pds has 6504 rows and 64 columns.

[passiondstats-blog]

Here’s a list of blogs that have already been created. I will update the list as more people sign up. Please reblog to share the list with other Courserians.

http://onemoresorrystatistic.tumblr.com/

http://higheredquestions.tumblr.com/

http://passiondstats.tumblr.com/

...

[higheredquestions]

Lecture 1

Use “categorical” over Qualitative? Is this common across the disciplines? 

[the-path-to-procrastination]

Qualitative is commonly used to refer to categorical variables; however this terminology is confusing as people think of qualitative research.

SPSS refers to these variables as Nominal / Ordinal.

R uses the term "factors", treating them as nominal variables, and ordered factors (as ordinal variables).

Stata uses the term categorical variables.

Btw - your comments are off, so I'm reblogging :)

[the-path-to-procrastination]

Quiz 6 result

Week 5 Quiz was a short but difficult one: I got 8/9 on the first attempt and 9/9 on the second one.

[the-path-to-procrastination]

Quiz 5 result

Week 4 Quiz 2 of 2 was long, but easier: 18/18 on the first attempt!

I think I'm doing well...

[the-path-to-procrastination]

Quiz 4 result

Week 4 Quiz 1 of 2 was a tough one. 17 questions to answer. I only got 16 right on the 1st attempt, but the second one gave me maximum points (17/17).

[the-path-to-procrastination]

Quiz 3 result

Week 3 Quiz was a tougher one - but I got 16/16 on the first attempt. Hey, I’m not bad at this!