SIOP 2020 Careers Projects
Kulas
Most recently updated Thursday January 28 2021 12:25:29 PM
SIOP 2020 Lale extension
Codebooks located at
5-year follow-up to Lale findings (which go through 2013; Lale (2015))
Lale
Lale Figure 1.
General linear downward effect has continued, \(R^2\) = 0.4654888. If we just focus on last few years, has this percentage of second job holders leveled off? (not sure best way analytically to test for this - polynomial? [8/3/19])
Note. Tried the 3-month floating average (“moving average.R”) but plot is extremely ragged (8/3/19)
Lale Figure 2a.
This is very ragged without geom_smooth
Smoothing across 3-month rolling average
Smoothing across 3-month rolling average
Smoothing across 3-month rolling average
Smoothing across 3-month rolling average
Smoothing across 3-month rolling average
Lale Figure 2C.
Lale Figure 2C.
Lale Figure 2C.
Maybe point out that it looks like White and Asian women who are predominantly driving the gender effect?
Lale Figure 2C.
Education with 3-month moving average estimate.
Age with 3-month moving average estimate.
Marital status with 3-month moving average estimate.
NOTE. Subsetting to only 2010 \(\rightarrow\) results in rMarkdown errors - will have to copy chunks separately for Lale (2015) graph continuation…
Recession data was taken from the National Bureau of Economic Research, which identified an 18-month United States recession lasting from December 2007 to June 2009. The categories of “Pre-recession”, “During”, and “Post-recession” are defined by this 18 month window.
Percentage second job holders per occupational category.
Overall percentages of second job-holders was seemingly not affected by the recession, with percentages being 6.04% Pre-recession, 5.81% During recession, and 5.4% Post-recession.
Regarding hours worked per week, first job reported figures ranged from 0 to 99, while self-reported other job earnings (from all other jobs [than the primary]) ranged from 0 to 99. We adjusted for seemingly out-of-range values by truncating cases to caps of 90 for primary job and 70 for “other” jobs. Summing the two variables resulted in total number of hours worked the previous week (across all jobs) estimates ranging from 1 to 130.
## `summarise()` regrouping output by 'second' (override with `.groups` argument)## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `summarise()` regrouping output by 'second' (override with `.groups` argument)
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `summarise()` regrouping output by 'second' (override with `.groups` argument)
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
Sleep variable is:
- lale$t010101 (Sleeping)
Socializing variable is:
- lale$t120101 (Socializing and communicating with others)
- lale$t120301 (Relaxing, thinking)
- lale$t120302 (Tobacco and drug use)
- lale$t120303 (Television and movies (not religious))
- lale$t120304 (Television (religious))
- lale$t120305 (Listening to the radio)
- lale$t120306 (Listening to/playing music (not radio))
- lale$t120307 (Playing games)
- lale$t120308 (Computer use for leisure (exc. Games))
- lale$t120309 (Arts and crafts as a hobby)
- lale$t120310 (Collecting as a hobby)
- lale$t120311 (Hobbies, except arts & crafts and collecting)
- lale$t120312 (Reading for personal interest)
- lale$t120313 (Writing for personal interest)
Exercise variable is:
- lale$t130101 (Doing aerobics)
- lale$t130102 (Playing baseball)
- lale$t130103 (Playing basketball)
- lale$t130104 (Biking)
- lale$t130105 (Playing billiards)
- lale$t130106 (Boating)
- lale$t130107 (Bowling)
- lale$t130108 (Climbing, spelunking, caving)
- lale$t130109 (Dancing)
- lale$t130110 (Participating in equestrian sports)
- lale$t130111 (Fencing)
- lale$t130112 (Fishing)
- lale$t130113 (Playing football)
- lale$t130114 (Golfing)
- lale$t130115 (Doing gymnastics)
- lale$t130116 (Hiking)
- lale$t130117 (Playing hockey)
- lale$t130118 (Hunting)
- lale$t130119 (Participating in martial arts)
- lale$t130120 (Playing racquet sports)
- lale$t130121 (Participating in rodeo competitions)
- lale$t130122 (Rollerblading)
- lale$t130123 (Playing rugby)
- lale$t130124 (Running)
- lale$t130125 (Skiing, ice skating, snowboarding)
- lale$t130126 (Playing soccer)
- lale$t130127 (Softball)
- lale$t130128 (Using cardiovascular equipment)
- lale$t130129 (Vehicle touring/racing)
- lale$t130130 (Playing volleyball)
- lale$t130131 (Walking)
- lale$t130132 (Participating in water sports)
- lale$t130133 (Weightlifting/strength training)
- lale$t130134 (Working out, unspecified)
- lale$t130135 (Wrestling)
- lale$t130136 (Doing yoga)
Looking at seasonal patterns (month variable is HRMONTH): Time spent on activities that are considered resources or recovery: Sleeping, Socializing, and Exercising
Looking at seasonal patterns (month variable is HRMONTH): Time spent Caring for HH members
Looking at Seasonal Patterns: Time spent in HH Activities
#First, see what class these things are and confirm category variables are factors
str(lale$second)## num [1:159512] 0 0 0 NA 0 0 0 0 0 0 ...str(lale$Year)## num [1:159512] 2003 2003 2003 2003 2003 ...str(lale$age)## chr [1:159512] "55 to 64 years" "25 to 54 years" "25 to 54 years" ...str(lale$race); lale$race <- factor(lale$race)## chr [1:159512] "Black" "White" "White" "Black" "White" "White" "White" ...str(lale$marry); lale$marry <- factor(lale$marry)## chr [1:159512] "Married" "Married" "Married" "Married" "Married" ...str(lale$educ3); lale$educ3 <- factor(lale$educ3)## chr [1:159512] "College or higher education" "Some college" "Some college" ...str(lale$income2); lale$income2 <- factor(lale$income2)## chr [1:159512] "$50,000 to $75,000" NA "$75,000 to $99,999" ...str(lale$tot.hrs)## int [1:159512] NA NA NA NA NA NA NA NA NA NA ...#Reorder the race and income factors
lale$race <- factor(lale$race, levels = c("White", "Black", "Asian","Multiracial", "Native American/Pacific Islander"))
levels(lale$race)## [1] "White" "Black"
## [3] "Asian" "Multiracial"
## [5] "Native American/Pacific Islander"lale$income2 <- factor(lale$income2, levels = c("$30,000 to $50,000", "Less than $15,000", "$15,000 to $30,000", "$50,000 to $75,000", "$75,000 to $99,999", "$100,000 to $149,999", "$150,000 and over"))
levels(lale$income2)## [1] "$30,000 to $50,000" "Less than $15,000" "$15,000 to $30,000"
## [4] "$50,000 to $75,000" "$75,000 to $99,999" "$100,000 to $149,999"
## [7] "$150,000 and over"#Check to see if we have people that fall into both single/multiple by IV
xtabs(~lale$second + lale$Year, data = merged)## lale$Year
## lale$second 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
## 0 11776 7740 7351 7235 6940 7210 7070 7046 6584 6478 5946
## 1 757 485 496 480 404 423 468 425 362 376 330
## lale$Year
## lale$second 2014 2015 2016 2017 2018
## 0 6061 5774 5448 5247 4922
## 1 343 306 305 327 269xtabs(~lale$second + lale$age, data = merged)## lale$age
## lale$second 16 to 24 years 25 to 54 years 55 to 64 years
## 0 8204 82576 18048
## 1 462 5063 1031xtabs(~lale$second + lale$race, data = merged)## lale$race
## lale$second White Black Asian Multiracial Native American/Pacific Islander
## 0 88943 13235 4334 1322 994
## 1 5480 777 147 107 45xtabs(~lale$second + lale$marry, data = merged)## lale$marry
## lale$second Married Single Widowed, divorced, or separated
## 0 61820 26737 20271
## 1 3464 1648 1444xtabs(~lale$second + lale$educ3, data = merged)## lale$educ3
## lale$second College or higher education High school graduate
## 0 42526 26066
## 1 3018 1141
## lale$educ3
## lale$second Less than high school Some college
## 0 8934 31302
## 1 246 2151xtabs(~lale$second + lale$income2, data = merged)## lale$income2
## lale$second $30,000 to $50,000 Less than $15,000 $15,000 to $30,000
## 0 20897 5101 14832
## 1 1368 266 928
## lale$income2
## lale$second $50,000 to $75,000 $75,000 to $99,999 $100,000 to $149,999
## 0 22052 17009 12887
## 1 1435 1081 707
## lale$income2
## lale$second $150,000 and over
## 0 9475
## 1 449xtabs(~lale$second + lale$tot.hrs, data = merged)## lale$tot.hrs
## lale$second 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
## 1 2 7 5 9 12 11 3 14 10 29 9 28 23 17 30 28 16
## lale$tot.hrs
## lale$second 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
## 1 35 17 62 28 49 35 53 63 38 32 55 29 114 27 69 48 58
## lale$tot.hrs
## lale$second 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
## 1 118 91 53 79 45 272 68 144 97 123 198 146 110 203 60 412 56
## lale$tot.hrs
## lale$second 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
## 1 210 72 87 268 148 67 84 43 414 39 59 48 69 141 43 27 48
## lale$tot.hrs
## lale$second 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
## 1 15 183 12 47 15 21 53 17 14 14 3 122 5 10 4 15 16
## lale$tot.hrs
## lale$second 86 87 88 90 91 92 93 94 95 96 97 98 99 100 104 108 109
## 1 4 2 5 22 2 9 1 4 11 7 1 2 1 19 1 3 1
## lale$tot.hrs
## lale$second 110 111 112 115 116 118 120 130
## 1 8 1 1 1 1 1 4 1logit <- glm(second ~ Year + gender + age + race + marry + educ3 + income2, data = lale, family = "binomial") # logistic regression
summary(logit) #here's the summary table##
## Call:
## glm(formula = second ~ Year + gender + age + race + marry + educ3 +
## income2, family = "binomial", data = lale)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.5312 -0.3827 -0.3381 -0.2874 2.9598
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 24.308117 5.810322 4.184 2.87e-05 ***
## Year -0.013283 0.002893 -4.591 4.41e-06 ***
## genderWomen 0.015898 0.026584 0.598 0.54980
## age25 to 54 years -0.089541 0.056835 -1.575 0.11515
## age55 to 64 years -0.165675 0.065613 -2.525 0.01157 *
## raceBlack -0.066199 0.041586 -1.592 0.11142
## raceAsian -0.659115 0.087988 -7.491 6.84e-14 ***
## raceMultiracial 0.236353 0.104378 2.264 0.02355 *
## raceNative American/Pacific Islander -0.237016 0.153826 -1.541 0.12336
## marrySingle 0.065713 0.036116 1.820 0.06883 .
## marryWidowed, divorced, or separated 0.218587 0.035742 6.116 9.61e-10 ***
## educ3High school graduate -0.652994 0.039239 -16.641 < 2e-16 ***
## educ3Less than high school -1.171472 0.073533 -15.931 < 2e-16 ***
## educ3Some college -0.169742 0.031622 -5.368 7.97e-08 ***
## income2Less than $15,000 -0.106511 0.069795 -1.526 0.12700
## income2$15,000 to $30,000 0.042984 0.044465 0.967 0.33370
## income2$50,000 to $75,000 -0.059184 0.039722 -1.490 0.13624
## income2$75,000 to $99,999 -0.136755 0.043898 -3.115 0.00184 **
## income2$100,000 to $149,999 -0.286607 0.050516 -5.674 1.40e-08 ***
## income2$150,000 and over -0.450071 0.059520 -7.562 3.98e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 47719 on 108486 degrees of freedom
## Residual deviance: 47010 on 108467 degrees of freedom
## (51025 observations deleted due to missingness)
## AIC: 47050
##
## Number of Fisher Scoring iterations: 6exp(coef(logit)) #these are supposed to be the odds ratios ????## (Intercept) Year
## 3.604799e+10 9.868047e-01
## genderWomen age25 to 54 years
## 1.016025e+00 9.143506e-01
## age55 to 64 years raceBlack
## 8.473217e-01 9.359446e-01
## raceAsian raceMultiracial
## 5.173088e-01 1.266621e+00
## raceNative American/Pacific Islander marrySingle
## 7.889783e-01 1.067920e+00
## marryWidowed, divorced, or separated educ3High school graduate
## 1.244318e+00 5.204852e-01
## educ3Less than high school educ3Some college
## 3.099104e-01 8.438823e-01
## income2Less than $15,000 income2$15,000 to $30,000
## 8.989649e-01 1.043921e+00
## income2$50,000 to $75,000 income2$75,000 to $99,999
## 9.425331e-01 8.721838e-01
## income2$100,000 to $149,999 income2$150,000 and over
## 7.508070e-01 6.375832e-01exp(cbind(OR = coef(logit), confint.default(logit))) #95% confidence intervals of standard errors## OR 2.5 % 97.5 %
## (Intercept) 3.604799e+10 4.084381e+05 3.181528e+15
## Year 9.868047e-01 9.812247e-01 9.924164e-01
## genderWomen 1.016025e+00 9.644432e-01 1.070367e+00
## age25 to 54 years 9.143506e-01 8.179647e-01 1.022094e+00
## age55 to 64 years 8.473217e-01 7.450719e-01 9.636037e-01
## raceBlack 9.359446e-01 8.626840e-01 1.015426e+00
## raceAsian 5.173088e-01 4.353658e-01 6.146748e-01
## raceMultiracial 1.266621e+00 1.032285e+00 1.554153e+00
## raceNative American/Pacific Islander 7.889783e-01 5.836170e-01 1.066602e+00
## marrySingle 1.067920e+00 9.949402e-01 1.146252e+00
## marryWidowed, divorced, or separated 1.244318e+00 1.160133e+00 1.334611e+00
## educ3High school graduate 5.204852e-01 4.819564e-01 5.620940e-01
## educ3Less than high school 3.099104e-01 2.683149e-01 3.579542e-01
## educ3Some college 8.438823e-01 7.931683e-01 8.978388e-01
## income2Less than $15,000 8.989649e-01 7.840305e-01 1.030748e+00
## income2$15,000 to $30,000 1.043921e+00 9.567953e-01 1.138980e+00
## income2$50,000 to $75,000 9.425331e-01 8.719369e-01 1.018845e+00
## income2$75,000 to $99,999 8.721838e-01 8.002805e-01 9.505475e-01
## income2$100,000 to $149,999 7.508070e-01 6.800312e-01 8.289489e-01
## income2$150,000 and over 6.375832e-01 5.673787e-01 7.164743e-01## APA Table for Output
library(gtsummary)## #Uighurtable <- tbl_regression(logit, exponentiate = TRUE, label = list(Year ~ "Year", gender ~ "Gender", age ~ "Age", race ~ "Race", marry ~ "Marital Status", educ3 ~ "Education level", income2 ~ "Income"))
table| Characteristic | OR1 | 95% CI1 | p-value |
|---|---|---|---|
| Year | 0.99 | 0.98, 0.99 | <0.001 |
| Gender | |||
| Men | — | — | |
| Women | 1.02 | 0.96, 1.07 | 0.5 |
| Age | |||
| 16 to 24 years | — | — | |
| 25 to 54 years | 0.91 | 0.82, 1.02 | 0.12 |
| 55 to 64 years | 0.85 | 0.75, 0.96 | 0.012 |
| Race | |||
| White | — | — | |
| Black | 0.94 | 0.86, 1.01 | 0.11 |
| Asian | 0.52 | 0.43, 0.61 | <0.001 |
| Multiracial | 1.27 | 1.03, 1.55 | 0.024 |
| Native American/Pacific Islander | 0.79 | 0.58, 1.05 | 0.12 |
| Marital Status | |||
| Married | — | — | |
| Single | 1.07 | 0.99, 1.15 | 0.069 |
| Widowed, divorced, or separated | 1.24 | 1.16, 1.33 | <0.001 |
| Education level | |||
| College or higher education | — | — | |
| High school graduate | 0.52 | 0.48, 0.56 | <0.001 |
| Less than high school | 0.31 | 0.27, 0.36 | <0.001 |
| Some college | 0.84 | 0.79, 0.90 | <0.001 |
| Income | |||
| $30,000 to $50,000 | — | — | |
| Less than $15,000 | 0.90 | 0.78, 1.03 | 0.13 |
| $15,000 to $30,000 | 1.04 | 0.96, 1.14 | 0.3 |
| $50,000 to $75,000 | 0.94 | 0.87, 1.02 | 0.14 |
| $75,000 to $99,999 | 0.87 | 0.80, 0.95 | 0.002 |
| $100,000 to $149,999 | 0.75 | 0.68, 0.83 | <0.001 |
| $150,000 and over | 0.64 | 0.57, 0.72 | <0.001 |
|
1
OR = Odds Ratio, CI = Confidence Interval
|
|||
save image
The odds ratios would be the odds of X depending on whether someone is a multiple job holder or not. The Confidence intervals should NOT include 1, if they do then they are not significant!
#References
Lale, Etienne. 2015. “Multiple Jobholding over the Past Two Decades : Monthly Labor Review: U.S. Bureau of Labor Statistics.” 2015. https://www.bls.gov/opub/mlr/2015/article/multiple-jobholding-over-the-past-two-decades.htm.