library(tidyverse)

## -- Attaching packages ----------------------------------------------------------------------------------------------- tidyverse 1.2.1 --

## v ggplot2 3.1.0       v purrr   0.2.5  
## v tibble  2.0.1       v dplyr   0.8.0.1
## v tidyr   0.8.2       v stringr 1.4.0  
## v readr   1.3.1       v forcats 0.4.0

## -- Conflicts -------------------------------------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

library(knitr)
library(kableExtra)

## 
## Attaching package: 'kableExtra'

## The following object is masked from 'package:dplyr':
## 
##     group_rows

Introduction

Choice of major in graduation has always been a major decision in every student’s life. Having a good estimate about the job and salary prospects in a major in future defintely helps students in making an informative decision about his/her grad choice. We have seen numerous articles advising which course to choose and which one would be in demand with prospective employers. ‘STEM’ which stands for Science Technology Engineering Mathematics, and has been highlighted to be a great choice in majors to find high paying jobs.
In this project, I’m researching to find whether the choice of STEM major in graduation has an impact on a graduate salary?

Data

Data collection:The recent grad list compiled by census.gov has provided us with a list of compiled data of grads during 2010-2012. There are observations made against 173 majors.
Cases: Each observation represents a major with data mentioning the number of grads who passed, their employement rate, salary ranges etc. There are totally 173 observations which each observation representing a major.
Variables:We will be studying two variables, the median salary which is a response variable and course major which is an independent variable also an explantory variable. This is an observation study since it’s derived from a data which is taken from an observation during a period of time.
The population of interest is the graduates of United States. And the findings from this analysis can be generalized for the entire population of interest i.e., the graduates.
Causality: We can use this data to establish a causality between the variables of interest. Since the data has been sampled randomly from a large population of data source.

Source: https://github.com/fivethirtyeight/data/tree/master/college-majors

grads <- read.csv("https://raw.githubusercontent.com/san123i/CUNY/master/Semester1/606/Final%20Project/grad-students.csv")
head(grads)%>%
  kable() %>%
  kable_styling() %>% scroll_box(width = "800px")

Major_code	Major	Major_category	Grad_total	Grad_sample_size	Grad_employed	Grad_full_time_year_round	Grad_unemployed	Grad_unemployment_rate	Grad_median	Grad_P25	Grad_P75	Nongrad_total	Nongrad_employed	Nongrad_full_time_year_round	Nongrad_unemployed	Nongrad_unemployment_rate	Nongrad_median	Nongrad_P25	Nongrad_P75	Grad_share	Grad_premium
5601	CONSTRUCTION SERVICES	Industrial Arts & Consumer Services	9173	200	7098	6511	681	0.0875434	75000	53000	110000	86062	73607	62435	3928	0.0506610	65000	47000	98000	0.0963196	0.1538462
6004	COMMERCIAL ART AND GRAPHIC DESIGN	Arts	53864	882	40492	29553	2482	0.0577559	60000	40000	89000	461977	347166	250596	25484	0.0683859	48000	34000	71000	0.1044198	0.2500000
6211	HOSPITALITY MANAGEMENT	Business	24417	437	18368	14784	1465	0.0738668	65000	45000	100000	179335	145597	113579	7409	0.0484229	50000	35000	75000	0.1198369	0.3000000
2201	COSMETOLOGY SERVICES AND CULINARY ARTS	Industrial Arts & Consumer Services	5411	72	3590	2701	316	0.0809012	47000	24500	85000	37575	29738	23249	1661	0.0528998	41600	29000	60000	0.1258782	0.1298077
2001	COMMUNICATION TECHNOLOGIES	Computers & Mathematics	9109	171	7512	5622	466	0.0584106	57000	40600	83700	53819	43163	34231	3389	0.0728003	52000	36000	78000	0.1447527	0.0961538
3201	COURT REPORTING	Law & Public Policy	1542	22	1008	860	0	0.0000000	75000	55000	120000	8921	6967	6063	518	0.0692051	50000	34000	75000	0.1473765	0.5000000

summary(grads)

##    Major_code                                     Major    
##  Min.   :1100   ACCOUNTING                           :  1  
##  1st Qu.:2403   ACTUARIAL SCIENCE                    :  1  
##  Median :3608   ADVERTISING AND PUBLIC RELATIONS     :  1  
##  Mean   :3880   AEROSPACE ENGINEERING                :  1  
##  3rd Qu.:5503   AGRICULTURAL ECONOMICS               :  1  
##  Max.   :6403   AGRICULTURE PRODUCTION AND MANAGEMENT:  1  
##                 (Other)                              :167  
##                    Major_category   Grad_total      Grad_sample_size
##  Engineering              :29     Min.   :   1542   Min.   :   22   
##  Education                :16     1st Qu.:  15284   1st Qu.:  314   
##  Humanities & Liberal Arts:15     Median :  37872   Median :  688   
##  Biology & Life Science   :14     Mean   : 127672   Mean   : 2251   
##  Business                 :13     3rd Qu.: 148255   3rd Qu.: 2528   
##  Health                   :12     Max.   :1184158   Max.   :21994   
##  (Other)                  :74                                       
##  Grad_employed    Grad_full_time_year_round Grad_unemployed
##  Min.   :  1008   Min.   :   770            Min.   :    0  
##  1st Qu.: 12659   1st Qu.:  9894            1st Qu.:  453  
##  Median : 28930   Median : 22523            Median : 1179  
##  Mean   : 94037   Mean   : 72861            Mean   : 3506  
##  3rd Qu.:109944   3rd Qu.: 80794            3rd Qu.: 3329  
##  Max.   :915341   Max.   :703347            Max.   :35718  
##                                                            
##  Grad_unemployment_rate  Grad_median        Grad_P25        Grad_P75     
##  Min.   :0.00000        Min.   : 47000   Min.   :24500   Min.   : 65000  
##  1st Qu.:0.02607        1st Qu.: 65000   1st Qu.:45000   1st Qu.: 93000  
##  Median :0.03665        Median : 75000   Median :50000   Median :108000  
##  Mean   :0.03934        Mean   : 76756   Mean   :52597   Mean   :112087  
##  3rd Qu.:0.04805        3rd Qu.: 90000   3rd Qu.:60000   3rd Qu.:130000  
##  Max.   :0.13851        Max.   :135000   Max.   :85000   Max.   :294000  
##                                                                          
##  Nongrad_total     Nongrad_employed  Nongrad_full_time_year_round
##  Min.   :   2232   Min.   :   1328   Min.   :    980             
##  1st Qu.:  20564   1st Qu.:  15914   1st Qu.:  11755             
##  Median :  68993   Median :  50092   Median :  38384             
##  Mean   : 214720   Mean   : 154554   Mean   : 120737             
##  3rd Qu.: 184971   3rd Qu.: 129179   3rd Qu.: 103629             
##  Max.   :2996892   Max.   :2253649   Max.   :1882507             
##                                                                  
##  Nongrad_unemployed Nongrad_unemployment_rate Nongrad_median  
##  Min.   :     0     Min.   :0.00000           Min.   : 37000  
##  1st Qu.:   880     1st Qu.:0.04198           1st Qu.: 48700  
##  Median :  3157     Median :0.05103           Median : 55000  
##  Mean   :  8486     Mean   :0.05395           Mean   : 58584  
##  3rd Qu.:  7409     3rd Qu.:0.06439           3rd Qu.: 65000  
##  Max.   :136978     Max.   :0.16091           Max.   :126000  
##                                                               
##   Nongrad_P25     Nongrad_P75       Grad_share       Grad_premium    
##  Min.   :25000   Min.   : 48000   Min.   :0.09632   Min.   :-0.0250  
##  1st Qu.:34000   1st Qu.: 72000   1st Qu.:0.26757   1st Qu.: 0.2308  
##  Median :38000   Median : 80000   Median :0.39875   Median : 0.3208  
##  Mean   :40078   Mean   : 84333   Mean   :0.40059   Mean   : 0.3285  
##  3rd Qu.:44000   3rd Qu.: 97000   3rd Qu.:0.49912   3rd Qu.: 0.4000  
##  Max.   :80000   Max.   :215000   Max.   :0.93117   Max.   : 1.6471  
##

Exploratory Data Analysis

isSTEM <- function(majorCategory){
    result <- grepl("science|math|engineering|technology|computer", majorCategory, ignore.case = T)
    result <- ifelse(result, !grepl("Social", majorCategory, ignore.case = T), result)
    return(ifelse(result, "STEM","NON-STEM"));
}

grads <- grads %>% mutate("Major_Type_STEM"=isSTEM(Major_category))

x <- grads %>% select(Major, Major_category, Major_Type_STEM)
#x[order(x$Major_Type_STEM),]
x[order(x$Major_Type_STEM),] %>%
  kable() %>%
  kable_styling()

	Major	Major_category	Major_Type_STEM
1	CONSTRUCTION SERVICES	Industrial Arts & Consumer Services	NON-STEM
2	COMMERCIAL ART AND GRAPHIC DESIGN	Arts	NON-STEM
3	HOSPITALITY MANAGEMENT	Business	NON-STEM
4	COSMETOLOGY SERVICES AND CULINARY ARTS	Industrial Arts & Consumer Services	NON-STEM
6	COURT REPORTING	Law & Public Policy	NON-STEM
7	MARKETING AND MARKETING RESEARCH	Business	NON-STEM
8	AGRICULTURE PRODUCTION AND MANAGEMENT	Agriculture & Natural Resources	NON-STEM
10	ADVERTISING AND PUBLIC RELATIONS	Communications & Journalism	NON-STEM
11	FILM VIDEO AND PHOTOGRAPHIC ARTS	Arts	NON-STEM
12	ELECTRICAL, MECHANICAL, AND PRECISION TECHNOLOGIES AND PRODUCTION	Industrial Arts & Consumer Services	NON-STEM
14	MASS MEDIA	Communications & Journalism	NON-STEM
15	TRANSPORTATION SCIENCES AND TECHNOLOGIES	Industrial Arts & Consumer Services	NON-STEM
17	MISCELLANEOUS BUSINESS & MEDICAL ADMINISTRATION	Business	NON-STEM
20	MISCELLANEOUS FINE ARTS	Arts	NON-STEM
21	CRIMINAL JUSTICE AND FIRE PROTECTION	Law & Public Policy	NON-STEM
22	BUSINESS MANAGEMENT AND ADMINISTRATION	Business	NON-STEM
23	CRIMINOLOGY	Social Science	NON-STEM
24	MANAGEMENT INFORMATION SYSTEMS AND STATISTICS	Business	NON-STEM
26	OPERATIONS LOGISTICS AND E-COMMERCE	Business	NON-STEM
27	GENERAL BUSINESS	Business	NON-STEM
28	MEDICAL TECHNOLOGIES TECHNICIANS	Health	NON-STEM
30	COMMUNICATIONS	Communications & Journalism	NON-STEM
31	ACTUARIAL SCIENCE	Business	NON-STEM
33	JOURNALISM	Communications & Journalism	NON-STEM
34	MEDICAL ASSISTING SERVICES	Health	NON-STEM
36	ACCOUNTING	Business	NON-STEM
37	FINE ARTS	Arts	NON-STEM
38	NURSING	Health	NON-STEM
41	MULTI/INTERDISCIPLINARY STUDIES	Interdisciplinary	NON-STEM
43	GENERAL AGRICULTURE	Agriculture & Natural Resources	NON-STEM
44	FORESTRY	Agriculture & Natural Resources	NON-STEM
45	LIBERAL ARTS	Humanities & Liberal Arts	NON-STEM
46	HUMAN SERVICES AND COMMUNITY ORGANIZATION	Psychology & Social Work	NON-STEM
47	VISUAL AND PERFORMING ARTS	Arts	NON-STEM
48	NATURAL RESOURCES MANAGEMENT	Agriculture & Natural Resources	NON-STEM
49	STUDIO ARTS	Arts	NON-STEM
50	FAMILY AND CONSUMER SCIENCES	Industrial Arts & Consumer Services	NON-STEM
51	PHYSICAL FITNESS PARKS RECREATION AND LEISURE	Industrial Arts & Consumer Services	NON-STEM
52	FINANCE	Business	NON-STEM
54	PLANT SCIENCE AND AGRONOMY	Agriculture & Natural Resources	NON-STEM
55	HUMAN RESOURCES AND PERSONNEL MANAGEMENT	Business	NON-STEM
56	INTERNATIONAL BUSINESS	Business	NON-STEM
57	COMPOSITION AND RHETORIC	Humanities & Liberal Arts	NON-STEM
58	DRAMA AND THEATER ARTS	Arts	NON-STEM
59	BUSINESS ECONOMICS	Business	NON-STEM
62	HEALTH AND MEDICAL ADMINISTRATIVE SERVICES	Health	NON-STEM
63	AGRICULTURAL ECONOMICS	Agriculture & Natural Resources	NON-STEM
65	GEOGRAPHY	Social Science	NON-STEM
68	INTERDISCIPLINARY SOCIAL SCIENCES	Social Science	NON-STEM
70	SOIL SCIENCE	Agriculture & Natural Resources	NON-STEM
71	PRE-LAW AND LEGAL STUDIES	Law & Public Policy	NON-STEM
77	EARLY CHILDHOOD EDUCATION	Education	NON-STEM
78	SOCIOLOGY	Social Science	NON-STEM
79	GENERAL SOCIAL SCIENCES	Social Science	NON-STEM
80	ANIMAL SCIENCES	Agriculture & Natural Resources	NON-STEM
81	TREATMENT THERAPY PROFESSIONS	Health	NON-STEM
82	MISCELLANEOUS AGRICULTURE	Agriculture & Natural Resources	NON-STEM
84	HUMANITIES	Humanities & Liberal Arts	NON-STEM
85	FOOD SCIENCE	Agriculture & Natural Resources	NON-STEM
88	SOCIAL PSYCHOLOGY	Psychology & Social Work	NON-STEM
91	ART HISTORY AND CRITICISM	Humanities & Liberal Arts	NON-STEM
92	MISCELLANEOUS HEALTH MEDICAL PROFESSIONS	Health	NON-STEM
93	GENERAL MEDICAL AND HEALTH SERVICES	Health	NON-STEM
94	INTERCULTURAL AND INTERNATIONAL STUDIES	Humanities & Liberal Arts	NON-STEM
95	NUTRITION SCIENCES	Health	NON-STEM
96	ECONOMICS	Social Science	NON-STEM
97	PHYSICAL AND HEALTH EDUCATION TEACHING	Education	NON-STEM
98	COMMUNITY AND PUBLIC HEALTH	Health	NON-STEM
100	THEOLOGY AND RELIGIOUS VOCATIONS	Humanities & Liberal Arts	NON-STEM
102	MISCELLANEOUS EDUCATION	Education	NON-STEM
104	PUBLIC ADMINISTRATION	Law & Public Policy	NON-STEM
105	ELEMENTARY EDUCATION	Education	NON-STEM
106	INDUSTRIAL AND ORGANIZATIONAL PSYCHOLOGY	Psychology & Social Work	NON-STEM
107	MILITARY TECHNOLOGIES	Industrial Arts & Consumer Services	NON-STEM
108	GENERAL EDUCATION	Education	NON-STEM
109	MUSIC	Arts	NON-STEM
110	ART AND MUSIC EDUCATION	Education	NON-STEM
111	LINGUISTICS AND COMPARATIVE LANGUAGE AND LITERATURE	Humanities & Liberal Arts	NON-STEM
113	ANTHROPOLOGY AND ARCHEOLOGY	Humanities & Liberal Arts	NON-STEM
114	SOCIAL WORK	Psychology & Social Work	NON-STEM
115	ENGLISH LANGUAGE AND LITERATURE	Humanities & Liberal Arts	NON-STEM
116	TEACHER EDUCATION: MULTIPLE LEVELS	Education	NON-STEM
118	PHARMACY PHARMACEUTICAL SCIENCES AND ADMINISTRATION	Health	NON-STEM
119	OTHER FOREIGN LANGUAGES	Humanities & Liberal Arts	NON-STEM
120	PSYCHOLOGY	Psychology & Social Work	NON-STEM
121	AREA ETHNIC AND CIVILIZATION STUDIES	Humanities & Liberal Arts	NON-STEM
126	HISTORY	Humanities & Liberal Arts	NON-STEM
127	MISCELLANEOUS SOCIAL SCIENCES	Social Science	NON-STEM
130	FRENCH GERMAN LATIN AND OTHER COMMON FOREIGN LANGUAGE STUDIES	Humanities & Liberal Arts	NON-STEM
131	SOCIAL SCIENCE OR HISTORY TEACHER EDUCATION	Education	NON-STEM
133	POLITICAL SCIENCE AND GOVERNMENT	Social Science	NON-STEM
134	INTERNATIONAL RELATIONS	Social Science	NON-STEM
137	MISCELLANEOUS PSYCHOLOGY	Psychology & Social Work	NON-STEM
139	SECONDARY TEACHER EDUCATION	Education	NON-STEM
141	UNITED STATES HISTORY	Humanities & Liberal Arts	NON-STEM
144	LANGUAGE AND DRAMA EDUCATION	Education	NON-STEM
146	PUBLIC POLICY	Law & Public Policy	NON-STEM
147	MATHEMATICS TEACHER EDUCATION	Education	NON-STEM
148	SCIENCE AND COMPUTER TEACHER EDUCATION	Education	NON-STEM
150	PHILOSOPHY AND RELIGIOUS STUDIES	Humanities & Liberal Arts	NON-STEM
151	SPECIAL NEEDS EDUCATION	Education	NON-STEM
158	LIBRARY SCIENCE	Education	NON-STEM
164	EDUCATIONAL PSYCHOLOGY	Psychology & Social Work	NON-STEM
168	COMMUNICATION DISORDERS SCIENCES AND SERVICES	Health	NON-STEM
169	COUNSELING PSYCHOLOGY	Psychology & Social Work	NON-STEM
170	CLINICAL PSYCHOLOGY	Psychology & Social Work	NON-STEM
171	HEALTH AND MEDICAL PREPARATORY PROGRAMS	Health	NON-STEM
172	SCHOOL STUDENT COUNSELING	Education	NON-STEM
173	EDUCATIONAL ADMINISTRATION AND SUPERVISION	Education	NON-STEM
5	COMMUNICATION TECHNOLOGIES	Computers & Mathematics	STEM
9	COMPUTER PROGRAMMING AND DATA PROCESSING	Computers & Mathematics	STEM
13	MECHANICAL ENGINEERING RELATED TECHNOLOGIES	Engineering	STEM
16	COMPUTER NETWORKING AND TELECOMMUNICATIONS	Computers & Mathematics	STEM
18	MISCELLANEOUS ENGINEERING TECHNOLOGIES	Engineering	STEM
19	INDUSTRIAL PRODUCTION TECHNOLOGIES	Engineering	STEM
25	COMPUTER ADMINISTRATION MANAGEMENT AND SECURITY	Computers & Mathematics	STEM
29	COMPUTER AND INFORMATION SYSTEMS	Computers & Mathematics	STEM
32	ELECTRICAL ENGINEERING TECHNOLOGY	Engineering	STEM
35	ENGINEERING TECHNOLOGIES	Engineering	STEM
39	INFORMATION SCIENCES	Computers & Mathematics	STEM
40	ARCHITECTURAL ENGINEERING	Engineering	STEM
42	NUCLEAR, INDUSTRIAL RADIOLOGY, AND BIOLOGICAL TECHNOLOGIES	Physical Sciences	STEM
53	PETROLEUM ENGINEERING	Engineering	STEM
60	ENGINEERING AND INDUSTRIAL MANAGEMENT	Engineering	STEM
61	COMPUTER SCIENCE	Computers & Mathematics	STEM
64	ENVIRONMENTAL SCIENCE	Biology & Life Science	STEM
66	MISCELLANEOUS ENGINEERING	Engineering	STEM
67	ECOLOGY	Biology & Life Science	STEM
69	ARCHITECTURE	Engineering	STEM
72	GENERAL ENGINEERING	Engineering	STEM
73	MULTI-DISCIPLINARY OR GENERAL SCIENCE	Physical Sciences	STEM
74	CIVIL ENGINEERING	Engineering	STEM
75	COMPUTER ENGINEERING	Engineering	STEM
76	MINING AND MINERAL ENGINEERING	Engineering	STEM
83	MECHANICAL ENGINEERING	Engineering	STEM
86	INDUSTRIAL AND MANUFACTURING ENGINEERING	Engineering	STEM
87	GEOLOGICAL AND GEOPHYSICAL ENGINEERING	Engineering	STEM
89	NAVAL ARCHITECTURE AND MARINE ENGINEERING	Engineering	STEM
90	MATHEMATICS AND COMPUTER SCIENCE	Computers & Mathematics	STEM
99	ELECTRICAL ENGINEERING	Engineering	STEM
101	OCEANOGRAPHY	Physical Sciences	STEM
103	BIOLOGICAL ENGINEERING	Engineering	STEM
112	MATERIALS ENGINEERING AND MATERIALS SCIENCE	Engineering	STEM
117	GEOLOGY AND EARTH SCIENCE	Physical Sciences	STEM
122	PHYSICAL SCIENCES	Physical Sciences	STEM
123	ATMOSPHERIC SCIENCES AND METEOROLOGY	Physical Sciences	STEM
124	CHEMICAL ENGINEERING	Engineering	STEM
125	AEROSPACE ENGINEERING	Engineering	STEM
128	APPLIED MATHEMATICS	Computers & Mathematics	STEM
129	STATISTICS AND DECISION SCIENCE	Computers & Mathematics	STEM
132	MATHEMATICS	Computers & Mathematics	STEM
135	ENVIRONMENTAL ENGINEERING	Engineering	STEM
136	MISCELLANEOUS BIOLOGY	Biology & Life Science	STEM
138	METALLURGICAL ENGINEERING	Engineering	STEM
140	GEOSCIENCES	Physical Sciences	STEM
142	ENGINEERING MECHANICS PHYSICS AND SCIENCE	Engineering	STEM
143	COGNITIVE SCIENCE AND BIOPSYCHOLOGY	Biology & Life Science	STEM
145	NUCLEAR ENGINEERING	Engineering	STEM
149	MICROBIOLOGY	Biology & Life Science	STEM
152	BOTANY	Biology & Life Science	STEM
153	BIOLOGY	Biology & Life Science	STEM
154	ASTRONOMY AND ASTROPHYSICS	Physical Sciences	STEM
155	CHEMISTRY	Physical Sciences	STEM
156	PHYSIOLOGY	Biology & Life Science	STEM
157	BIOMEDICAL ENGINEERING	Engineering	STEM
159	MOLECULAR BIOLOGY	Biology & Life Science	STEM
160	PHARMACOLOGY	Biology & Life Science	STEM
161	ZOOLOGY	Biology & Life Science	STEM
162	PHYSICS	Physical Sciences	STEM
163	NEUROSCIENCE	Biology & Life Science	STEM
165	BIOCHEMICAL SCIENCES	Biology & Life Science	STEM
166	GENETICS	Biology & Life Science	STEM
167	MATERIALS SCIENCE	Engineering	STEM

The below plot shows that majority of the graduates fall under ‘Education’ major type where as the least are under ‘interdiscriplinary’ and ‘Agriculture & Natural Resources’ category. Also, there are only 4 major categories which fall under STEM i.e. ‘Biology & Life Sciences’, ‘Computers & Mathematics’, ‘Engineering’ and ‘Physical Sciences’.

1. Number of graduates in each major

Let’s do an analysis on the number of graduates in each major.

ggplot(grads, aes(x=reorder(Major_category, Grad_total), y = Grad_total, fill=Major_Type_STEM)) + geom_bar(stat="identity") + theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))+ labs( title = 'Graduates by major category type')

list <- aggregate(Grad_total ~ Major_category, grads, sum)
ggplot(list, aes(x=reorder(Major_category, Grad_total), y=Grad_total, label=Grad_total, label.size=.25)) + geom_col() + theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))+ labs(title = '# of Graduates by major category type') + 
  geom_text(size = 3, position = position_stack(vjust = 0.5), color='white')

list <- aggregate(Grad_total ~ Major_Type_STEM, grads, sum)
ggplot(list, aes(x=reorder(Major_Type_STEM, Grad_total), y=Grad_total, label=Grad_total, fill=Major_Type_STEM)) + geom_col() + theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))+ labs(title = '# of Graduates by STEM type') + 
  geom_text(size = 3, position = position_stack(vjust = 0.5), color='white')

# label=sum(Grad_total)

2. Salary data analysis by category

The below plot shows the grad median salary in each major category.

ggplot(grads, aes(Major_category, Grad_median, fill=Major_Type_STEM)) + geom_boxplot()  +  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))

ggplot(grads, aes(x=Major, y=Grad_median, shape=Major_Type_STEM, colour=Major_Type_STEM)) + geom_point()  + scale_shape_manual(values=c(19, 2))

3. Average ‘p25’, ‘p75’ and ‘Median’ salary of each category.

ggplot(grads, aes(x=reorder(Major_category, Grad_P25), y = Grad_P25, fill=Major_Type_STEM)) + geom_bar(stat="identity") +labs(title = 'a. 25Perc of Grad salary by Major_category') +  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))

ggplot(grads, aes(x=reorder(Major_category, Grad_P75), y = Grad_P75, fill=Major_Type_STEM)) + geom_bar(stat="identity") +labs(title = 'b. 75Perc of Grad salary by Major_category') +  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))

ggplot(grads, aes(x=reorder(Major_category, Grad_median), y = Grad_median, fill=Major_Type_STEM)) + geom_bar(stat="identity") +labs(title = 'c. Median of Grad salary by Major_category') +  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))

4. Grad Median salaries by STEM or Non-STEM

ggplot(grads, aes(Major_Type_STEM, Grad_median, fill=Major_Type_STEM)) + geom_boxplot()  +  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5))

Inference

We are using Analysis of Variance (ANOVA) method to dervie an inference from the Hypothesis. Below are the conditions we evaluated and found to be satisfied.
a. Independence of cases - The cases provided as per data are independent in nature
b. Distribution of Residuals are Normal - The below qqnorm graph displays that residuals are normally distributed without major outliers
c. Homogentiy of variances

Let’s start our inference with a Null Hypothesis test
H0 <- The median salary for an average graduate in STEM and Non-STEM categories are same
HA <- The median salary for an average graduate in STEM and non-STEM categories are not same.

grads_anova <- aov(grads$Grad_median ~ grads$Major_Type_STEM)
summary(grads_anova)

##                        Df    Sum Sq   Mean Sq F value Pr(>F)    
## grads$Major_Type_STEM   1 1.644e+10 1.644e+10    85.8 <2e-16 ***
## Residuals             171 3.276e+10 1.916e+08                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

plot(grads_anova)

Since the P value is less than .05, we reject the null hypothesis (H0), and can conclude that average graduate salary for STEM and non-STEM category students is not equal. And from the geom plot, we can identify that Median salary from STEM category students is significantly higher than non-STEM categories.

Conclusion

Using the given data, we were able to do an exploratory analysis and identify a clear pattern/trend that STEM course major students have a median salary higher than non-grad students. And using the ANOVA analysis, we were able to confirm that grad median salaries for both STEM and Non-STEM grad student are NOT equals. Therefore, we can clearly state that grad median salaries are higher for STEM students.
Future research: I believe there are few other factors which can influence the grad salary such as market conditions, number of open job opportunities, type of economies i.e., a developed, emerging or under developed. If I can get access to these datasets, then an even more realistic and robust model can be designed.

GradStudent_STEM_Impact_606FinalProject

SantoshCheruku

April 28, 2019