GGR 276 Lab 2 Part 2
Scarlett Rakowska
2026-07-31
Part B1. Sampling Data
Import Data
If your R Markdown is NOT in the same folder as your
data, please set your working directory using setwd()
first. Here is an example if you are using
Medusa
setwd("\\medusa\StudentWork\(Your UTOR ID)\GGR276\Lab1")
and if you are using your own computer
‘setwd(C:/)’. You will need to change the code to reflect your personal
directory. Then load your data. When loading in the data try to get your
‘Data variable name’ short and easy to type! If you change it from what
is listed in the rmarkdown instructions you will need to change it.
We are loading the same aviation facilities dataset as Assignment 1.
Once the dataset is loaded, you will need to investigate the dataset to determine if you have any NA values. It is also recommended you determine the descriptive statistics before proceeding. We are building on the skills you learned in lab 1. This includes summary statistics and filtering your data. In statistics, each step builds on the previous.
setwd("~/UTM/Year 5/Summer/Term 2/GGR276/Assignments/Assignment 2")
aviation_facilities <- read.csv("aviationfacilities.csv", sep = ",", header = TRUE)View your dataframe and determine the distribution of the all the variables in this dataframe; are you missing any data?:
#ToDo
summary(aviation_facilities)## EFF_DATE SITE_NO SITE_TYPE_CODE STATE_CODE
## Length :19707 Min. : 103 Length :19707 Length :19707
## N.unique : 1 1st Qu.: 5251 N.unique : 6 N.unique : 60
## N.blank : 0 Median :13294 N.blank : 0 N.blank : 147
## Min.nchar: 10 Mean :15633 Min.nchar: 1 Min.nchar: 0
## Max.nchar: 10 3rd Qu.:23404 Max.nchar: 1 Max.nchar: 2
## Max. :90970
##
## ARPT_ID CITY COUNTRY_CODE REGION_CODE
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique :19707 N.unique : 8044 N.unique : 16 N.unique : 10
## N.blank : 0 N.blank : 0 N.blank : 0 N.blank : 147
## Min.nchar: 3 Min.nchar: 3 Min.nchar: 2 Min.nchar: 0
## Max.nchar: 4 Max.nchar: 35 Max.nchar: 2 Max.nchar: 3
##
##
## ADO_CODE STATE_NAME COUNTY_NAME COUNTY_ASSOC_STATE
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 23 N.unique : 60 N.unique : 1802 N.unique : 73
## N.blank : 3753 N.blank : 147 N.blank : 0 N.blank : 0
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 3 Min.nchar: 2
## Max.nchar: 3 Max.nchar: 17 Max.nchar: 21 Max.nchar: 2
##
##
## ARPT_NAME OWNERSHIP_TYPE_CODE FACILITY_USE_CODE LAT_DEG
## Length :19707 Length :19707 Length :19707 Min. : 5.00
## N.unique :18487 N.unique : 6 N.unique : 2 1st Qu.:33.00
## N.blank : 0 N.blank : 0 N.blank : 0 Median :39.00
## Min.nchar: 2 Min.nchar: 2 Min.nchar: 2 Mean :38.48
## Max.nchar: 50 Max.nchar: 2 Max.nchar: 2 3rd Qu.:42.00
## Max. :76.00
##
## LAT_MIN LAT_SEC LAT_HEMIS LAT_DECIMAL
## Min. : 0.00 Min. : 0.00 Length :19707 Min. :-14.33
## 1st Qu.:15.00 1st Qu.:14.36 N.unique : 2 1st Qu.: 33.97
## Median :30.00 Median :29.94 N.blank : 0 Median : 39.06
## Mean :29.68 Mean :29.56 Min.nchar: 1 Mean : 38.98
## 3rd Qu.:44.00 3rd Qu.:44.89 Max.nchar: 1 3rd Qu.: 42.77
## Max. :59.00 Max. :60.00 Max. : 76.53
##
## LONG_DEG LONG_MIN LONG_SEC LONG_HEMIS
## Min. : 14.00 Min. : 0.00 Min. : 0.00 Length :19707
## 1st Qu.: 82.00 1st Qu.:14.00 1st Qu.:14.30 N.unique : 2
## Median : 91.00 Median :29.00 Median :29.96 N.blank : 0
## Mean : 94.98 Mean :29.13 Mean :29.72 Min.nchar: 1
## 3rd Qu.:102.00 3rd Qu.:44.00 3rd Qu.:45.00 Max.nchar: 1
## Max. :177.00 Max. :59.00 Max. :59.99
##
## LONG_DECIMAL SURVEY_METHOD_CODE ELEV ELEV_METHOD_CODE
## Min. :-177.38 Length :19707 Min. : -223 Length :19707
## 1st Qu.:-102.13 N.unique : 1 1st Qu.: 275 N.unique : 3
## Median : -91.81 N.blank : 0 Median : 755 N.blank : 466
## Mean : -94.93 Min.nchar: 1 Mean : 1189 Min.nchar: 0
## 3rd Qu.: -82.48 Max.nchar: 1 3rd Qu.: 1273 Max.nchar: 1
## Max. : 174.11 Max. :12442
##
## MAG_VARN MAG_HEMIS MAG_VARN_YEAR TPA
## Min. : 0.000 Length :19707 Min. :1904 Min. : 130.0
## 1st Qu.: 3.000 N.unique : 3 1st Qu.:1985 1st Qu.: 800.0
## Median : 7.000 N.blank : 5314 Median :1985 Median : 800.0
## Mean : 8.384 Min.nchar: 0 Mean :1993 Mean : 883.4
## 3rd Qu.:13.000 Max.nchar: 1 3rd Qu.:1995 3rd Qu.: 968.5
## Max. :31.000 Max. :2030 Max. :4500.0
## NAs :5314 NAs :5438 NAs :19169
## CHART_NAME DIST_CITY_TO_AIRPORT DIRECTION_CODE ACREAGE
## Length :19707 Min. : 0.000 Length :19707 Min. : 0.0
## N.unique : 58 1st Qu.: 1.000 N.unique : 17 1st Qu.: 10.0
## N.blank : 437 Median : 3.000 N.blank : 207 Median : 52.0
## Min.nchar: 0 Mean : 3.888 Min.nchar: 0 Mean : 290.8
## Max.nchar: 26 3rd Qu.: 5.000 Max.nchar: 3 3rd Qu.: 195.0
## Max. :96.000 Max. :55000.0
## NAs :152 NAs :10046
## RESP_ARTCC_ID COMPUTER_ID ARTCC_NAME FSS_ON_ARPT_FLAG
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 35 N.unique : 35 N.unique : 35 N.unique : 3
## N.blank : 0 N.blank : 0 N.blank : 0 N.blank : 8898
## Min.nchar: 3 Min.nchar: 3 Min.nchar: 4 Min.nchar: 0
## Max.nchar: 4 Max.nchar: 3 Max.nchar: 25 Max.nchar: 1
##
##
## FSS_ID FSS_NAME PHONE_NO TOLL_FREE_NO
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 86 N.unique : 86 N.unique : 19 N.unique : 19
## N.blank : 0 N.blank : 0 N.blank :18797 N.blank : 85
## Min.nchar: 3 Min.nchar: 4 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 4 Max.nchar: 17 Max.nchar: 12 Max.nchar: 14
##
##
## ALT_FSS_ID ALT_FSS_NAME ALT_TOLL_FREE_NO NOTAM_ID
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 4 N.unique : 4 N.unique : 4 N.unique : 1849
## N.blank :19489 N.blank :19489 N.blank :19489 N.blank :13968
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 3 Max.nchar: 9 Max.nchar: 14 Max.nchar: 4
##
##
## NOTAM_FLAG ACTIVATION_DATE ARPT_STATUS FAR_139_TYPE_CODE
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 3 N.unique : 1057 N.unique : 3 N.unique : 11
## N.blank : 6623 N.blank : 929 N.blank : 0 N.blank :19189
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 1 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 7 Max.nchar: 2 Max.nchar: 5
##
##
## FAR_139_CARRIER_SER_CODE ARFF_CERT_TYPE_DATE NASP_CODE
## Length :19707 Length :19707 Length :19707
## N.unique : 3 N.unique : 142 N.unique : 119
## N.blank :19189 N.blank :19189 N.blank :16220
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 7 Max.nchar: 7
##
##
## ASP_ANLYS_DTRM_CODE CUST_FLAG LNDG_RIGHTS_FLAG JOINT_USE_FLAG
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 5 N.unique : 3 N.unique : 3 N.unique : 3
## N.blank : 463 N.blank :15474 N.blank :15502 N.blank :15326
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 13 Max.nchar: 1 Max.nchar: 1 Max.nchar: 1
##
##
## MIL_LNDG_FLAG INSPECT_METHOD_CODE INSPECTOR_CODE LAST_INSPECTION
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 3 N.unique : 6 N.unique : 4 N.unique : 1510
## N.blank :15242 N.blank : 4044 N.blank : 0 N.blank :14066
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 1 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 1 Max.nchar: 1 Max.nchar: 10
##
##
## LAST_INFO_RESPONSE FUEL_TYPES AIRFRAME_REPAIR_SER_CODE
## Length :19707 Length :19707 Length :19707
## N.unique : 3540 N.unique : 78 N.unique : 4
## N.blank : 5936 N.blank :15809 N.blank :13660
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 10 Max.nchar: 18 Max.nchar: 5
##
##
## PWR_PLANT_REPAIR_SER BOTTLED_OXY_TYPE BULK_OXY_TYPE LGT_SKED
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 4 N.unique : 5 N.unique : 5 N.unique : 3
## N.blank :13661 N.blank :15695 N.blank :15743 N.blank :13078
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 5 Max.nchar: 8 Max.nchar: 8 Max.nchar: 7
##
##
## BCN_LGT_SKED TWR_TYPE_CODE SEG_CIRCLE_MKR_FLAG BCN_LENS_COLOR
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 3 N.unique : 6 N.unique : 4 N.unique : 8
## N.blank :15310 N.blank : 0 N.blank : 5006 N.blank :15208
## Min.nchar: 0 Min.nchar: 4 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 7 Max.nchar: 11 Max.nchar: 3 Max.nchar: 3
##
##
## LNDG_FEE_FLAG MEDICAL_USE_FLAG ARPT_PSN_SOURCE POSITION_SRC_DATE
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 3 N.unique : 3 N.unique : 19 N.unique : 5342
## N.blank :10951 N.blank :16697 N.blank : 6053 N.blank : 6345
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 1 Max.nchar: 16 Max.nchar: 10
##
##
## ARPT_ELEV_SOURCE ELEVATION_SRC_DATE CONTR_FUEL_AVBL TRNS_STRG_BUOY_FLAG
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 19 N.unique : 5445 N.unique : 3 N.unique : 3
## N.blank : 8428 N.blank : 8637 N.blank :19589 N.blank :19576
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 16 Max.nchar: 10 Max.nchar: 1 Max.nchar: 1
##
##
## TRNS_STRG_HGR_FLAG TRNS_STRG_TIE_FLAG OTHER_SERVICES WIND_INDCR_FLAG
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 3 N.unique : 3 N.unique : 441 N.unique : 4
## N.blank :16733 N.blank :14939 N.blank :16483 N.blank : 4676
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 1 Max.nchar: 54 Max.nchar: 3
##
##
## ICAO_ID MIN_OP_NETWORK USER_FEE_FLAG CTA
## Length :19707 Length :19707 Length :19707 Length :19707
## N.unique : 2706 N.unique : 2 N.unique : 2 N.unique : 45
## N.blank :17002 N.blank : 0 N.blank :19644 N.blank :19516
## Min.nchar: 0 Min.nchar: 1 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 4 Max.nchar: 1 Max.nchar: 26 Max.nchar: 4
##
## Now that we have looked at the dataframe and know some of the summary statistics, we want to filter the data so that we only keep facilities located in continental USA. We also want to remove NA values from Acreage and Elevation so we can use those variables later for sampling and inferential statistics.
# Filtering the data to subset the dataset:
# We only want facilities where STATE_CODE is in continental USA
# We also remove records with NA values for ACREAGE or ELEV (needed later in the lab)
#ToDo
filtered_states <- subset(
aviation_facilities,
(STATE_CODE %in% c("AL","AZ","AR","CA","CO","CT","DE","FL","GA",
"ID","IL","IN","IA","KS","KY","LA","ME","MD",
"MA","MI","MN","MS","MO","MT","NE","NV","NH",
"NJ","NM","NY","NC","ND","OH","OK","OR","PA",
"RI","SC","SD","TN","TX","UT","VT","VA","WA","DC",
"WV","WI","WY")) & !is.na(ACREAGE) & !is.na(ELEV)
)
nrow(filtered_states)## [1] 9350table(filtered_states$SITE_TYPE_CODE)##
## A B C G H U
## 7822 8 94 24 1337 65summary(filtered_states$ACREAGE)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0 10.0 51.0 286.6 190.0 55000.0summary(filtered_states$ELEV)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -210.0 331.1 803.5 1316.5 1420.0 9933.5Now that we have filtered to continental USA, we will focus only on 3 facility types: Airports (A), Heliports (H), and Seaplane Bases (C).
# Use SITE_TYPE_CODE to filter the dataset to only include "A", "H", and "C".
#ToDo
filtered_states<-subset(filtered_states,(SITE_TYPE_CODE %in% c("A","H","C")) & !is.na(ACREAGE) & !is.na(ELEV))
table(filtered_states$SITE_TYPE_CODE)##
## A C H
## 7822 94 1337Simple Random Sampling
First, we will randomly sample 100 records. from the filtered dataset.
random_sample <- filtered_states[sample(1:nrow(filtered_states), size = 100, replace = FALSE), ]
nrow(random_sample)## [1] 100Stratified Sampling
Next, we will try stratified sampling using dplyr package. In this exercise, we are only interested in facilities where Elevation is greater than 1000 (ELEV > 1000). Therefore, we create a subset first.
# Create a subset consisting of Elevation >= 1000
subset_facilities <- subset(filtered_states, ELEV >= 1000)
nrow(subset_facilities)## [1] 3569table(subset_facilities$SITE_TYPE_CODE)##
## A C H
## 3199 16 354# Stratified Sampling
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionstratified_sample <- subset_facilities %>%
group_by(SITE_TYPE_CODE) %>%
sample_n(size = 10, replace = FALSE) %>%
ungroup()
nrow(stratified_sample)## [1] 30table(stratified_sample$SITE_TYPE_CODE)##
## A C H
## 10 10 10(4 marks) Q7 Now, it is your turn.
Please create a stratified sample consisting of facilities where Elevation is above the 1st quartile elevation, Acreage is above the 1st quartile acreage, and Remove NA values from Elevation and Acreage (already done in filtered_states).
For the stratified sample, you will be using the SITE_TYPE_CODE category. For each category, make the sample size equal to 20. To make the result reproducible, please add code set.seed(123) prior to your sampling.
Please indicate the number of observations in your subset and the number of observations in your stratified sample. You can use summary to find the 1st quartiles.
summary(filtered_states)## EFF_DATE SITE_NO SITE_TYPE_CODE STATE_CODE
## Length :9253 Min. : 103 Length :9253 Length :9253
## N.unique : 1 1st Qu.: 5293 N.unique : 3 N.unique : 49
## N.blank : 0 Median :13083 N.blank : 0 N.blank : 0
## Min.nchar: 10 Mean :13812 Min.nchar: 1 Min.nchar: 2
## Max.nchar: 10 3rd Qu.:22890 Max.nchar: 1 Max.nchar: 2
## Max. :42289
##
## ARPT_ID CITY COUNTRY_CODE REGION_CODE
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique :9253 N.unique :5132 N.unique : 1 N.unique : 8
## N.blank : 0 N.blank : 0 N.blank : 0 N.blank : 0
## Min.nchar: 3 Min.nchar: 3 Min.nchar: 2 Min.nchar: 3
## Max.nchar: 4 Max.nchar: 33 Max.nchar: 2 Max.nchar: 3
##
##
## ADO_CODE STATE_NAME COUNTY_NAME COUNTY_ASSOC_STATE
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 21 N.unique : 49 N.unique :1654 N.unique : 49
## N.blank :1106 N.blank : 0 N.blank : 0 N.blank : 0
## Min.nchar: 0 Min.nchar: 4 Min.nchar: 3 Min.nchar: 2
## Max.nchar: 3 Max.nchar: 17 Max.nchar: 19 Max.nchar: 2
##
##
## ARPT_NAME OWNERSHIP_TYPE_CODE FACILITY_USE_CODE LAT_DEG
## Length :9253 Length :9253 Length :9253 Min. :24.00
## N.unique :8846 N.unique : 5 N.unique : 2 1st Qu.:33.00
## N.blank : 0 N.blank : 0 N.blank : 0 Median :38.00
## Min.nchar: 2 Min.nchar: 2 Min.nchar: 2 Mean :37.67
## Max.nchar: 50 Max.nchar: 2 Max.nchar: 2 3rd Qu.:42.00
## Max. :48.00
##
## LAT_MIN LAT_SEC LAT_HEMIS LAT_DECIMAL
## Min. : 0.00 Min. : 0.00 Length :9253 Min. :24.56
## 1st Qu.:15.00 1st Qu.:14.51 N.unique : 1 1st Qu.:33.78
## Median :30.00 Median :30.00 N.blank : 0 Median :38.58
## Mean :29.81 Mean :29.60 Min.nchar: 1 Mean :38.17
## 3rd Qu.:45.00 3rd Qu.:44.60 Max.nchar: 1 3rd Qu.:42.33
## Max. :59.00 Max. :59.99 Max. :49.00
##
## LONG_DEG LONG_MIN LONG_SEC LONG_HEMIS
## Min. : 67.00 Min. : 0.00 Min. : 0.00 Length :9253
## 1st Qu.: 82.00 1st Qu.:15.00 1st Qu.:14.67 N.unique : 1
## Median : 91.00 Median :29.00 Median :30.16 N.blank : 0
## Mean : 93.18 Mean :29.26 Mean :29.95 Min.nchar: 1
## 3rd Qu.:100.00 3rd Qu.:44.00 3rd Qu.:45.25 Max.nchar: 1
## Max. :124.00 Max. :59.00 Max. :59.99
##
## LONG_DECIMAL SURVEY_METHOD_CODE ELEV ELEV_METHOD_CODE
## Min. :-124.56 Length :9253 Min. :-210.0 Length :9253
## 1st Qu.:-100.91 N.unique : 1 1st Qu.: 332.6 N.unique : 3
## Median : -91.90 N.blank : 0 Median : 805.0 N.blank : 106
## Mean : -93.68 Min.nchar: 1 Mean :1320.5 Min.nchar: 0
## 3rd Qu.: -82.57 Max.nchar: 1 3rd Qu.:1430.0 Max.nchar: 1
## Max. : -67.01 Max. :9933.5
##
## MAG_VARN MAG_HEMIS MAG_VARN_YEAR TPA
## Min. : 0.000 Length :9253 Min. :1965 Min. : 260.0
## 1st Qu.: 3.000 N.unique : 3 1st Qu.:1985 1st Qu.: 800.0
## Median : 7.000 N.blank :1593 Median :1990 Median : 800.0
## Mean : 7.792 Min.nchar: 0 Mean :1994 Mean : 872.8
## 3rd Qu.:12.000 Max.nchar: 1 3rd Qu.:2000 3rd Qu.: 957.8
## Max. :21.000 Max. :2030 Max. :4200.0
## NAs :1593 NAs :1622 NAs :8773
## CHART_NAME DIST_CITY_TO_AIRPORT DIRECTION_CODE ACREAGE
## Length :9253 Min. : 0.000 Length :9253 Min. : 0.0
## N.unique : 38 1st Qu.: 2.000 N.unique : 17 1st Qu.: 10.0
## N.blank : 83 Median : 3.000 N.blank : 41 Median : 52.0
## Min.nchar: 0 Mean : 3.913 Min.nchar: 0 Mean : 288.9
## Max.nchar: 15 3rd Qu.: 5.000 Max.nchar: 3 3rd Qu.: 194.0
## Max. :68.000 Max. :55000.0
## NAs :28
## RESP_ARTCC_ID COMPUTER_ID ARTCC_NAME FSS_ON_ARPT_FLAG
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 22 N.unique : 22 N.unique : 22 N.unique : 2
## N.blank : 0 N.blank : 0 N.blank : 0 N.blank :2991
## Min.nchar: 3 Min.nchar: 3 Min.nchar: 5 Min.nchar: 0
## Max.nchar: 3 Max.nchar: 3 Max.nchar: 14 Max.nchar: 1
##
##
## FSS_ID FSS_NAME PHONE_NO TOLL_FREE_NO
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 56 N.unique : 56 N.unique : 2 N.unique : 1
## N.blank : 0 N.blank : 0 N.blank :9178 N.blank : 0
## Min.nchar: 3 Min.nchar: 4 Min.nchar: 0 Min.nchar: 14
## Max.nchar: 3 Max.nchar: 17 Max.nchar: 12 Max.nchar: 14
##
##
## ALT_FSS_ID ALT_FSS_NAME ALT_TOLL_FREE_NO NOTAM_ID
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 1 N.unique : 1 N.unique : 1 N.unique :1555
## N.blank :9253 N.blank :9253 N.blank :9253 N.blank :4776
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 0 Max.nchar: 0 Max.nchar: 0 Max.nchar: 3
##
##
## NOTAM_FLAG ACTIVATION_DATE ARPT_STATUS FAR_139_TYPE_CODE
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 3 N.unique :1027 N.unique : 3 N.unique : 11
## N.blank :2068 N.blank : 495 N.blank : 0 N.blank :8784
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 1 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 7 Max.nchar: 2 Max.nchar: 5
##
##
## FAR_139_CARRIER_SER_CODE ARFF_CERT_TYPE_DATE NASP_CODE
## Length :9253 Length :9253 Length :9253
## N.unique : 3 N.unique : 131 N.unique : 114
## N.blank :8784 N.blank :8784 N.blank :6086
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 7 Max.nchar: 7
##
##
## ASP_ANLYS_DTRM_CODE CUST_FLAG LNDG_RIGHTS_FLAG JOINT_USE_FLAG
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 5 N.unique : 3 N.unique : 3 N.unique : 3
## N.blank : 80 N.blank :5674 N.blank :5662 N.blank :5597
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 13 Max.nchar: 1 Max.nchar: 1 Max.nchar: 1
##
##
## MIL_LNDG_FLAG INSPECT_METHOD_CODE INSPECTOR_CODE LAST_INSPECTION
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 3 N.unique : 6 N.unique : 4 N.unique :1270
## N.blank :5672 N.blank : 984 N.blank : 0 N.blank :4466
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 1 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 1 Max.nchar: 1 Max.nchar: 10
##
##
## LAST_INFO_RESPONSE FUEL_TYPES AIRFRAME_REPAIR_SER_CODE
## Length :9253 Length :9253 Length :9253
## N.unique :2280 N.unique : 62 N.unique : 4
## N.blank :3966 N.blank :5842 N.blank :4940
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 10 Max.nchar: 18 Max.nchar: 5
##
##
## PWR_PLANT_REPAIR_SER BOTTLED_OXY_TYPE BULK_OXY_TYPE LGT_SKED
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 4 N.unique : 5 N.unique : 5 N.unique : 3
## N.blank :4937 N.blank :5725 N.blank :5783 N.blank :4923
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 5 Max.nchar: 8 Max.nchar: 8 Max.nchar: 7
##
##
## BCN_LGT_SKED TWR_TYPE_CODE SEG_CIRCLE_MKR_FLAG BCN_LENS_COLOR
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 3 N.unique : 5 N.unique : 4 N.unique : 8
## N.blank :5748 N.blank : 0 N.blank :1419 N.blank :5699
## Min.nchar: 0 Min.nchar: 4 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 7 Max.nchar: 11 Max.nchar: 3 Max.nchar: 3
##
##
## LNDG_FEE_FLAG MEDICAL_USE_FLAG ARPT_PSN_SOURCE POSITION_SRC_DATE
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 3 N.unique : 3 N.unique : 17 N.unique :3822
## N.blank :3839 N.blank :8652 N.blank :2655 N.blank :2827
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 1 Max.nchar: 16 Max.nchar: 10
##
##
## ARPT_ELEV_SOURCE ELEVATION_SRC_DATE CONTR_FUEL_AVBL TRNS_STRG_BUOY_FLAG
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 18 N.unique :3714 N.unique : 3 N.unique : 3
## N.blank :3616 N.blank :3692 N.blank :9188 N.blank :9159
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 16 Max.nchar: 10 Max.nchar: 1 Max.nchar: 1
##
##
## TRNS_STRG_HGR_FLAG TRNS_STRG_TIE_FLAG OTHER_SERVICES WIND_INDCR_FLAG
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique : 3 N.unique : 3 N.unique : 427 N.unique : 4
## N.blank :6464 N.blank :4788 N.blank :6314 N.blank : 902
## Min.nchar: 0 Min.nchar: 0 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 1 Max.nchar: 1 Max.nchar: 54 Max.nchar: 3
##
##
## ICAO_ID MIN_OP_NETWORK USER_FEE_FLAG CTA
## Length :9253 Length :9253 Length :9253 Length :9253
## N.unique :2136 N.unique : 2 N.unique : 2 N.unique : 35
## N.blank :7118 N.blank : 0 N.blank :9190 N.blank :9118
## Min.nchar: 0 Min.nchar: 1 Min.nchar: 0 Min.nchar: 0
## Max.nchar: 4 Max.nchar: 1 Max.nchar: 26 Max.nchar: 4
##
## subset_facilities<-subset(filtered_states,ELEV>332.6 & ACREAGE>10)
nrow(subset_facilities)## [1] 5181set.seed(123)
stratified_sample<-subset_facilities %>% group_by(SITE_TYPE_CODE) %>% sample_n(size=20,replace=FALSE) %>% ungroup()
nrow(stratified_sample)## [1] 60# ToDo
# Students can use summary() to identify the 1st quartiles (25th percentile)Compare Sampling Technique
(4 marks) Q8 What is the difference between simple random sampling and stratified sampling? In this case, would you use simple random sampling or stratified sampling?
Type your response here Simple random sampling is choosing a specific number of observations from a dataset and choosing data values randomly.Every data value has the same probability of being selected. Stratified sampling is when a number of observations is chosen from a dataset, the values are separated into groups depending on the data, and then random values are selected from each of these groups. This ensures that the overall sample is easily represented. In this case, stratified sampling would be the better option because of the three types of facility types. If we sample this filtered data randomly, there is a good chance that one of the facility types will be underrepresented.
Part B2. Inferential Statistics
Two-sample difference of mean test
(2 marks) Q9 Does the mean elevation differ between AIRPORT and HELIPORT facilities? Write the null hypothesis and alternate hypothesis?
summary(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="H"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 424.0 772.5 1465.0 2458.9 4255.0 6480.0summary(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="A"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 335.0 757.5 941.5 1703.2 2249.4 4606.4Type your response here Heliport facilities have a greater mean than Airport facilities, with a mean of 2458.9 and 1703.2 respectively.The null hypothesis is that μ airport=μ heliport and the alternate hypothesis is μairport!=μheliport.
(3 marks) Q10 What would happen if you used a parametric difference of mean test on non-parametric data?
Type your response here These two kinds of data are different. Parametric tests look at the actual values, while non parametric tests look at the ranking of these values. If you decide to do a parametric difference of mean test on non parametric data, the results will be unreliable because you are not looking at the values itself.
(4 marks) Q11 Please select an appropriate statistical test, and provide justifications by visualizing the distribution, using the stratified sample subset.
Type your response here Looking at the data, I would use a two sample t test because even in the stratified sample, it is a fairly large sample size, and we are also comparing two independent variables to each other, as airports and heliports are separate types of facilites. Looking at the boxplot, the distributions of data are similar, as both of them have the 1st quartile closer to the lower whisker than the higher whisker.
# ToDo
boxplot(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="A"],main="Airport Elevation",xlab="Elevation (m)")
boxplot(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="H"],main="Airport Elevation",xlab="Elevation (m)")
(6 marks) Q12 Please conduct the statistic test and interpret the results. Create 2 vectors (airport and heliport), each containing elevation values from each facility type. Then use a non-parametric difference of means test.
#TODO (2 marks)
airport<-(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="A"])
heliport<-(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="H"])
t.test(airport,heliport)##
## Welch Two Sample t-test
##
## data: airport and heliport
## t = -1.2994, df = 32.719, p-value = 0.2029
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1939.4166 427.9466
## sample estimates:
## mean of x mean of y
## 1703.215 2458.950wilcox.test(airport,heliport)##
## Wilcoxon rank sum exact test
##
## data: airport and heliport
## W = 165.5, p-value = 0.3583
## alternative hypothesis: true location shift is not equal to 0Type the statistic results and interpretation here
Three-sample Mean Difference Test
(8 marks) Q13 Please follow the same procedure above to test whether the 1st quartile elevation differs between facility types. You need to decide which three-or-more sample test to perform.
Type your response here:
#ToDo
# Visualize Distribution (1 mark)
Q1_airport<-quantile(airport,0.25)
View(Q1_airport)
Q1_heliport<-quantile(heliport,0.25)
View(Q1_heliport)
summary(heliport)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 424.0 772.5 1465.0 2458.9 4255.0 6480.0seaplanebase<-(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="C"])
Q1_seaplanebase<-quantile(seaplanebase,0.25)
View(Q1_seaplanebase)
boxplot(stratified_sample$ELEV[stratified_sample$SITE_TYPE_CODE=="C"],main="Airport Elevation",xlab="Elevation (m)")
class(seaplanebase)## [1] "numeric"kruskal.test(list(airport,heliport,seaplanebase))##
## Kruskal-Wallis rank sum test
##
## data: list(airport, heliport, seaplanebase)
## Kruskal-Wallis chi-squared = 3.7042, df = 2, p-value = 0.1569#ToDo
# Test (2 marks) using the filtered dataset (A/H/C)#Type 1 and Type 2 error (4 marks) Q14 Discuss what the Type 1 and Type 2 errors would be for both your 2-sample and 3-sample difference of means tests. Why is it important to understand Type 1 and Type 2 errors?
For my two sample and 3 sample tests, type 1 errors would state that I rejected the null hypothesis when it was true. In other words, I concluded that the difference in means was significant when it was actually equal to each other. Type 2 errors state that I accepted the null hypothesis when it was false. In other words, I concluded that there was no change in the means of the variables when in reality there was. It is important to understand Type 1 and 2 errors because if you understand what these statements mean in the context of the question, it will minimize error and ensure a more accurate concluding statement.