Chapter 2 Using existing data files
You have some data that you are eager to analyze using R. Before you can analyze the data, you must first read the data into R. This chapter describes how you can read common types of data files into R.
2.1 Working directories
I would recommend against using absolute pathway with the setwd() function for your works. One main reason is that other people don’t have the same pathways as yours, and it makes replicating codes in other people’s devices extremely cumbersome. There are two ways I find more efficient and productive than just setting up your directory with setwd().
- Open RStudio by doublie-clicking the
.Rfile. - Create an R project and open RStudio by double-clicking the
.Rprojfile.
Both methods will automatically set up your working directory to the folder where your .R or .Rproj file exists. Then you use relative pathway to navigate to other folders and files.
If you have followed the section on downloading dmus2 datasets, you might already have a data folder. You can put this data folder under your project folder, whatever you may call it. For the sake of demonstration, let’s create a folder named dmur and put this data folder under dmur. You can start create necessary .R and .Rproj files as follows. Essentially, you will have the following structures under thedmur folder.
- dmur ## this is your project or root folder
|
- data ## this is data folder that contains `dmus2` datasets.
|
- many datasets ... ## these are `dmus2` datasets
|
- dmur.R ## You can double-click this to open RStudio
|
- dmur.Rproj ## You can also double-click this to open RStudio[To add how to create .R or .Rproj files]
2.2 Reading data files into R
There are several data files that you can read and import into R. This chapter begins by illustrating how you can read several common types of datasets into R. As you would expect, you can certainly do this in multiple ways; however I will show the ways I think simplest and most efficient.
In previous section 1.8, we have downloaded several datasets into respective folders under a data folder of your current working directory. It should look like the following structure.
- Your current working directory
|
_ data
|
_ dmus2 ## <<== This is where dmus2 datasets were downloaded ==>>
|
_ who ## <<== This is where who datasets were downloaded ==>>When we specify a path later, we will use this folder structure. If we want to import a dataset from dmus2 folder, we will specify the file in this format ./data/dmus2/filename.
2.2.1 CSV and text data files
Comma-separated values files, also known as CSV files, are a common format for storing raw data. They have a filename extension of .csv. As the name suggests, CSV files use commas to separate the variables (columns) of the data. Data files in CSV format often include the names of the variables in the first row, also separated by commas.
We illustrate this below by using a momkid1.csv data file, which includes identification number of mom, month, day and year of mom’s birthday, and kid’s birthday.
While there are several ways to import a raw data file, we use read_csv() function from the readr package. As we use this function probably once at this time when we import datasets, we write readr::read_csv() to indicate the function from readr package without loading the whole package. We will use this pattern in the whole book whenever we need one particular function from a package at one time.
momkid1 <- readr::read_csv("data/dmus2/momkid1.csv")## Rows: 4 Columns: 5## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): kidbday
## dbl (4): momid, momm, momd, momy##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.momkid1## # A tibble: 4 × 5
## momid momm momd momy kidbday
## <dbl> <dbl> <dbl> <dbl> <chr>
## 1 1 11 28 1972 1/5/1998
## 2 2 4 3 1973 4/11/2002
## 3 3 6 13 1968 5/15/1996
## 4 4 1 5 1960 1/4/2004The imported dataset can be displayed in the console just by typing in the assigned object name momkid1.
Text files are also quite popular in storing simple raw data. In the example below, the dentists1.txt data file contains five rows of data regarding five dentists. The four variables reflect the name of the dentist, the years she or he has been practicing, whether she or he works full time, and whether she or he recommends Quaddent gum.
dentists1 <- readr::read_csv("data/dmus2/dentists1.txt")## Rows: 5 Columns: 4## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): name
## dbl (3): years, fulltime, recom##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.dentists1## # A tibble: 5 × 4
## name years fulltime recom
## <chr> <dbl> <dbl> <dbl>
## 1 Y. Don Uflossmore 7.25 0 1
## 2 Olive Tu'Drill 10.2 1 1
## 3 Isaac O'Yerbreath 32.8 1 1
## 4 Ruth Canaale 22 1 1
## 5 Mike Avity 8.5 0 0See ?read_csv for more details on their optional arguments.
2.2.1.1 Exercises
We have downloaded a few mortality datasets from WHO. Try to import the following datasets without looking at the answer:
popwhich contains reference population* and live birth data.country_codeswhich contains country codes
2.2.1.2 Solutions
As you might realize, there is no file extension for these files. Try opening these files with a text editor. You will see that these files are stored in comma-separated format. Hence, we will read using readr::read_csv()
pop <- readr::read_csv("data/who/pop")## Rows: 9719 Columns: 33## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): SubDiv, Frmat
## dbl (31): Country, Admin1, Year, Sex, Pop1, Pop2, Pop3, Pop4, Pop5, Pop6, Po...##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.pop## # A tibble: 9,719 × 33
## Country Admin1 SubDiv Year Sex Frmat Pop1 Pop2 Pop3 Pop4 Pop5
## <dbl> <dbl> <chr> <dbl> <dbl> <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1060 NA <NA> 1980 1 07 137100 3400 15800 NA NA
## 2 1060 NA <NA> 1980 2 07 159000 4000 18400 NA NA
## 3 1125 NA <NA> 1955 1 02 5051500 150300 543400 NA NA
## 4 1125 NA <NA> 1955 2 02 5049400 145200 551000 NA NA
## 5 1125 NA <NA> 1956 1 02 5353700 158700 576600 NA NA
## 6 1125 NA <NA> 1956 2 02 5351400 153600 584800 NA NA
## 7 1125 NA <NA> 1957 1 02 5403000 160300 580800 NA NA
## 8 1125 NA <NA> 1957 2 02 5392000 155300 589400 NA NA
## 9 1125 NA <NA> 1958 1 02 5506900 162800 592000 NA NA
## 10 1125 NA <NA> 1958 2 02 5494400 157800 600100 NA NA
## # … with 9,709 more rows, and 22 more variables: Pop6 <dbl>, Pop7 <dbl>,
## # Pop8 <dbl>, Pop9 <dbl>, Pop10 <dbl>, Pop11 <dbl>, Pop12 <dbl>, Pop13 <dbl>,
## # Pop14 <dbl>, Pop15 <dbl>, Pop16 <dbl>, Pop17 <dbl>, Pop18 <dbl>,
## # Pop19 <dbl>, Pop20 <dbl>, Pop21 <dbl>, Pop22 <dbl>, Pop23 <dbl>,
## # Pop24 <dbl>, Pop25 <dbl>, Pop26 <dbl>, Lb <dbl>country_codes <- readr::read_csv("data/who/country_codes")## Rows: 227 Columns: 2## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): name
## dbl (1): country##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.country_codes## # A tibble: 227 × 2
## country name
## <dbl> <chr>
## 1 1010 Algeria
## 2 1020 Angola
## 3 1025 Benin
## 4 1030 Botswana
## 5 1035 Burkina Faso
## 6 1040 Burundi
## 7 1045 Cameroon
## 8 1060 Cape Verde
## 9 1070 Central African Republic
## 10 1080 Chad
## # … with 217 more rows2.2.2 Excel spreadsheets
You frequently receive data files stored as Excel spreadsheets. For example, the spreadsheet file named dentists.xls contains information about five dentists. You can import Excel spreadsheets into R by using read_excel() from the readxl package.
dentists <- readxl::read_excel("data/dmus2/dentists.xls")
dentists## # A tibble: 5 × 4
## name years fulltime recom
## <chr> <dbl> <dbl> <dbl>
## 1 Y. Don Uflossmore 7.25 0 1
## 2 Olive Tu'Drill 10.2 1 1
## 3 Isaac O'Yerbreath 32.8 1 1
## 4 Ruth Canaale 22 1 1
## 5 Mike Avity 8.5 0 0As we did with dentists.xls, let’s import another file named dentists2.xls using the read_excel() function.
dentists2 <- readxl::read_excel("data/dmus2/dentists2.xls")
dentists2## # A tibble: 4 × 2
## name years
## <chr> <dbl>
## 1 I. Sue Yoo 3
## 2 A. Dewey 8
## 3 B. Cheetem 11
## 4 C. Howe 21Only 2 variables and 4 observations were imported. These are the names of lawyers. If you open the file with Excel or a spreadsheet program, you will realize that there are three sheets in the file, namely lawyers, dentists, and Sheet2. When there are multiple sheets in an Excel file, the default behavior for the read_excel() function is to import the first sheet. As such, the results from the list command above is showing a listing of lawyers because the first sheet contained information about lawyers.
We can import the contents of the sheet named dentists by adding the optional argument sheet = "dentists", as illustrated below. This tells the function that it should specifically import the sheet named dentists.
dentists2 <- readxl::read_excel("data/dmus2/dentists2.xls", sheet = "dentists")
dentists2## # A tibble: 5 × 4
## name years fulltime recom
## <chr> <dbl> <dbl> <dbl>
## 1 Y. Don Uflossmore 7.25 0 1
## 2 Olive Tu'Drill 10.2 1 1
## 3 Isaac O'Yerbreath 32.8 1 1
## 4 Ruth Canaale 22 1 1
## 5 Mike Avity 8.5 0 0As we would expect, the imported dataset has 4 variables and 5 observations which matches the contents of the dentists sheet.
Another file named dentist3.xls has additional information stored along with the data. In particular, column E contains notes about the dentists, and the last row, row 7, contains column totals for some of the variables. Let’s try to import this excel file using read_excel().
dentists3 <- readxl::read_excel("data/dmus2/dentists3.xls")## New names:
## * `` -> ...5dentists3## # A tibble: 6 × 5
## name years fulltime recom ...5
## <chr> <dbl> <dbl> <dbl> <chr>
## 1 Y. Don Uflossmore 7.25 0 1 <NA>
## 2 Olive Tu'Drill 10.2 1 1 Good with children
## 3 Isaac O'Yerbreath 32.8 1 1 <NA>
## 4 Ruth Canaale 22 1 1 <NA>
## 5 Mike Avity 8.5 0 0 Has evening appointments
## 6 Total 80.8 3 4 <NA>As you can see, the function imported all 5 variables and 6 observations including the Total row. We are only interested in the first four columns and five rows. Let’s include the range optional argument. In Excel’s terms, we want the data from cell A1 to cell D6. The codes below illustrate how to import a specified cell range into R.
dentists3 <- readxl::read_excel("data/dmus2/dentists3.xls", range = "A1:D6")
dentists3## # A tibble: 5 × 4
## name years fulltime recom
## <chr> <dbl> <dbl> <dbl>
## 1 Y. Don Uflossmore 7.25 0 1
## 2 Olive Tu'Drill 10.2 1 1
## 3 Isaac O'Yerbreath 32.8 1 1
## 4 Ruth Canaale 22 1 1
## 5 Mike Avity 8.5 0 0The console output now indicates that we have successfully imported the 4 variables and 5 observations we wanted.
See ?read_excel for more details on their optional arguments.
2.2.2.1 Exercises
Import the following excel spreadsheets:
list_ctry_yrs_21June2021.xlsxinwhofolder that contains A list of countries-years available for the mortality and population data.miliary_tb.xlsxinothersfolder that contains information about patients with miliary tuberculosis.
2.2.2.2 Solutions
Let’s import list_ctry_yrs_21June2021.xlsx using readxl::read_excel().
list_ctry_yrs <- readxl::read_excel("data/who/list_ctry_yrs_21June2021.xlsx")## New names:
## * `` -> ...2
## * `` -> ...3
## * `` -> ...4
## * `` -> ...5
## * `` -> ...6
## * ...list_ctry_yrs## # A tibble: 5,919 × 8
## `World Health Organization Mortal… ...2 ...3 ...4 ...5 ...6 ...7 ...8
## <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
## 1 "Availability of countries-years … <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 2 "In column \"ICD\", \"Icd7\" da… <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 3 "Consult file \"Documentation.do… <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 4 "Cells in orange indicate the ad… <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 5 "If you wish to be alerted of the… <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 6 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 7 "Country" name Admi… SubD… Year List Icd Upda…
## 8 "4005" Alban… <NA> <NA> 1987 09B Icd9 <NA>
## 9 "4005" Alban… <NA> <NA> 1988 09B Icd9 <NA>
## 10 "4005" Alban… <NA> <NA> 1989 09B Icd9 <NA>
## # … with 5,909 more rowsAs you can see, this is not correct yet. Apparently, there are some texts in several rows above the data. If you open this file with a spreadsheet program and count them, you know there are exactly seven rows that we need to skip.
We specify another argument called skip in the function to illustrate this.
list_ctry_yrs <- readxl::read_excel("data/who/list_ctry_yrs_21June2021.xlsx", skip = 7)list_ctry_yrs## # A tibble: 5,912 × 8
## Country name Admin1 SubDiv Year List Icd Update
## <chr> <chr> <chr> <lgl> <chr> <chr> <chr> <chr>
## 1 4005 Albania <NA> NA 1987 09B Icd9 <NA>
## 2 4005 Albania <NA> NA 1988 09B Icd9 <NA>
## 3 4005 Albania <NA> NA 1989 09B Icd9 <NA>
## 4 4005 Albania <NA> NA 1992 09B Icd9 <NA>
## 5 4005 Albania <NA> NA 1993 09B Icd9 <NA>
## 6 4005 Albania <NA> NA 1994 09B Icd9 <NA>
## 7 4005 Albania <NA> NA 1995 09B Icd9 <NA>
## 8 4005 Albania <NA> NA 1996 09B Icd9 <NA>
## 9 4005 Albania <NA> NA 1997 09B Icd9 <NA>
## 10 4005 Albania <NA> NA 1998 09B Icd9 <NA>
## # … with 5,902 more rowsSimilary, we can do this by specifying an exact range.
list_ctry_yrs <- readxl::read_excel("data/who/list_ctry_yrs_21June2021.xlsx",
range = "A8:H5920")list_ctry_yrs## # A tibble: 5,912 × 8
## Country name Admin1 SubDiv Year List Icd Update
## <chr> <chr> <chr> <lgl> <chr> <chr> <chr> <chr>
## 1 4005 Albania <NA> NA 1987 09B Icd9 <NA>
## 2 4005 Albania <NA> NA 1988 09B Icd9 <NA>
## 3 4005 Albania <NA> NA 1989 09B Icd9 <NA>
## 4 4005 Albania <NA> NA 1992 09B Icd9 <NA>
## 5 4005 Albania <NA> NA 1993 09B Icd9 <NA>
## 6 4005 Albania <NA> NA 1994 09B Icd9 <NA>
## 7 4005 Albania <NA> NA 1995 09B Icd9 <NA>
## 8 4005 Albania <NA> NA 1996 09B Icd9 <NA>
## 9 4005 Albania <NA> NA 1997 09B Icd9 <NA>
## 10 4005 Albania <NA> NA 1998 09B Icd9 <NA>
## # … with 5,902 more rowsFor the miliary tuberculosis, it’s straightforward to import this dataset.
miliary_tb <- readxl::read_excel("./data/others/miliary_tb.xlsx")
miliary_tb## # A tibble: 81 × 34
## age gender diagnosis duration cough sputum fiver nightsweat fatigue
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 4 2 1 10 1 0 0 0 0
## 2 43 1 1 60 1 1 1 1 0
## 3 20 2 1 45 0 0 1 0 0
## 4 22 2 1 30 0 0 1 0 0
## 5 67 1 1 40 1 0 1 0 0
## 6 6 1 1 13 0 0 1 0 0
## 7 21 2 1 14 1 1 1 0 0
## 8 22 1 1 13 0 0 1 0 0
## 9 30 1 1 120 1 1 1 0 0
## 10 36 1 1 3 1 1 1 0 0
## # … with 71 more rows, and 25 more variables: emaciation <dbl>, nausea <dbl>,
## # vomit <dbl>, chestpain <dbl>, shortbreath <dbl>, headache <dbl>, WBC <dbl>,
## # RBC <dbl>, HGB <dbl>, PLT <dbl>, Perc Neutr <dbl>, ALT <dbl>, AST <dbl>,
## # TP <dbl>, ALB <dbl>, GLB <dbl>, TBIL <dbl>, DBIL <dbl>, IBIL <dbl>,
## # GGT <dbl>, BUN <dbl>, CREA <dbl>, ESR <dbl>, CRP <dbl>, ADA <dbl>2.2.3 Stata datasets
Let’s read the same dataset dentists using Stata .dta format. We use the read_dta() function from the haven package.
dentists <- haven::read_dta("data/dmus2/dentists.dta")
dentists## # A tibble: 5 × 4
## name years fulltime recom
## <chr> <dbl> <dbl> <dbl>
## 1 Y. Don Uflossmore 7.25 0 1
## 2 Olive Tu'Drill 10.2 1 1
## 3 Isaac O'Yerbreath 32.8 1 1
## 4 Ruth Canaale 22 1 1
## 5 Mike Avity 8.5 0 0As you can see, we successfully read this dataset. The console output shows the information from the five dentists: their names, the years they have been practicing, whether they work full time, and whether they recommend Quaddent gum.
In addition to reading datasets from your computer, you can also read datasets stored on remote web servers. For example, dentists.dta is located on the Stata Press website, and you can use it by specifying its web URL address, as follows.
dentists <- haven::read_dta("https://www.stata-press.com/data/dmus2/dentists.dta")Often our dataset might be enormous. Let’s pretend that dentists.dta contains many variables, and we are only interested in importing just the variables name and years. We can import just these variables from dentists.dta, as shown below.
dentists <- haven::read_dta("data/dmus2/dentists.dta",
col_select = c("name", "years"))
dentists## # A tibble: 5 × 2
## name years
## <chr> <dbl>
## 1 Y. Don Uflossmore 7.25
## 2 Olive Tu'Drill 10.2
## 3 Isaac O'Yerbreath 32.8
## 4 Ruth Canaale 22
## 5 Mike Avity 8.52.2.3.1 Exercises
Import all the Stata datasets under ucla folder. Check the folder using dir().
2.2.3.2 Solutions
## list ulca folder
dir("data/ucla/")## [1] "doctor_stata_dm.dta" "hsb2.csv"
## [3] "hsb2.dta" "hsb2.xls"
## [5] "patient_pt1_stata_dm.dta" "patient_pt2_stata_dm.dta"doctor_stata_dm <- haven::read_dta("data/ucla/doctor_stata_dm.dta")
doctor_stata_dm## # A tibble: 40 × 5
## docid experience school lawsuits medicaid
## <chr> <dbl> <chr> <dbl> <dbl>
## 1 1-1 25 average 3 0.606
## 2 1-11 10 top 0 0.605
## 3 1-21 21 average 3 0.483
## 4 1-22 22 top 3 0.483
## 5 1-33 16 top 0 0.584
## 6 1-48 23 average 3 0.219
## 7 1-57 21 average 1 0.405
## 8 1-58 21 average 1 0.405
## 9 1-72 24 average 4 0.522
## 10 1-73 14 average 1 0.522
## # … with 30 more rowspatient_pt1_stata_dm <- haven::read_dta("data/ucla/patient_pt1_stata_dm.dta")
patient_pt1_stata_dm## # A tibble: 120 × 25
## hospital hospid docid dis_date tumorsize co2 pain wound mobility ntumors
## <chr> <dbl> <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 UCLA 1 1-1 2009-09-06 68.0 1.53 4 4 2 0
## 2 UCLA 1 1-1 2011-01-07 64.7 1.68 2 3 2 0
## 3 UCLA 1 1-1 2010-09-04 86.4 1.45 3 3 2 0
## 4 UCLA 1 1-1 2010-06-25 53.4 1.57 3 4 2 0
## 5 UCLA 1 1-1 2009-07-01 51.7 1.42 4 5 2 0
## 6 UCLA 1 1-1 2009-03-06 78.9 1.71 3 4 2 0
## 7 UCLA 1 1-1 2010-04-15 62.9 1.54 4 4 3 2
## 8 UCLA 1 1-11 2010-07-25 73.2 1.45 4 5 9 9
## 9 UCLA 1 1-11 2009-07-12 81.2 1.55 5 5 9 0
## 10 UCLA 1 1-11 2009-08-19 61.3 1.49 7 7 8 2
## # … with 110 more rows, and 15 more variables: nmorphine <dbl>,
## # remission <dbl>, lungcapacity <dbl>, age <dbl>, married <dbl>,
## # familyhx <chr>, smokinghx <chr>, sex <chr>, cancerstage <chr>,
## # lengthofstay <dbl>, wbc <chr>, rbc <dbl>, bmi <dbl>, test1 <dbl>,
## # test2 <dbl>patient_pt2_stata_dm <- haven::read_dta("data/ucla/patient_pt2_stata_dm.dta")
patient_pt2_stata_dm## # A tibble: 111 × 24
## hospital hospid docid dis_date tumorsize co2 pain wound mobility
## <chr> <dbl> <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 "Cedars-Sinai" 3 3-227 2009-10-01 69.8 1.53 6 5 5
## 2 "Cedars-Sinai" 3 3-227 2010-02-18 68.0 1.69 8 4 5
## 3 "Cedars-Sinai" 3 3-227 2009-06-30 65.1 1.56 7 4 6
## 4 "Cedars-Sinai" 3 3-227 2009-11-15 71.4 -98 9 5 6
## 5 "Cedars-Sinai" 3 3-227 2010-02-17 69.5 1.68 4 5 6
## 6 "Cedars-Sinai" 3 3-227 2009-12-22 89.7 1.89 5 6 5
## 7 " Cedars-Sinai" 3 3-227 2010-04-15 73.1 1.78 6 4 5
## 8 "Cedars-Sinai" 3 3-241 2010-04-17 80.6 1.53 3 7 6
## 9 "Cedars-Sinai" 3 3-241 2010-10-12 55.1 1.71 5 7 6
## 10 "Cedars-Sinai" 3 3-241 2009-10-30 61.3 1.78 8 7 6
## # … with 101 more rows, and 15 more variables: ntumors <dbl>, remission <dbl>,
## # lungcapacity <dbl>, age <dbl>, married <dbl>, familyhx <chr>,
## # smokinghx <chr>, sex <chr>, cancerstage <chr>, lengthofstay <dbl>,
## # wbc <chr>, rbc <dbl>, bmi <dbl>, test1 <dbl>, test2 <dbl>2.2.4 SPSS datasets
The read_spss() function from the haven package can import IBM SPSS Statistics .sav files. The examples below illustrate this by using an IBM SPSS Statistics file named dentlab.sav.
dentlab <- haven::read_spss("./data/dmus2/dentlab.sav")
dentlab## # A tibble: 5 × 4
## name years fulltime recom
## <chr> <dbl> <dbl+lbl> <dbl+lbl>
## 1 Y. Don Uflossmore 7.25 0 [part time] 1 [recommend]
## 2 Olive Tu'Drill 10.2 1 [full time] 1 [recommend]
## 3 Isaac O'Yerbreath 32.8 1 [full time] 1 [recommend]
## 4 Ruth Canaale 22 1 [full time] 1 [recommend]
## 5 Mike Avity 8.5 0 [part time] 0 [do not recommend]See ?labelled_spss for how labelled variables in SPSS are handled by the haven package in R.
Similar to read_dta(), we can import just the variables name and years from dentlab.sav, as shown below.
dentlab <- haven::read_spss("data/dmus2/dentlab.sav",
col_select = c("name", "years"))
dentlab## # A tibble: 5 × 2
## name years
## <chr> <dbl>
## 1 Y. Don Uflossmore 7.25
## 2 Olive Tu'Drill 10.2
## 3 Isaac O'Yerbreath 32.8
## 4 Ruth Canaale 22
## 5 Mike Avity 8.5Another function read_sav() can read both .sav and .zsav files from IBM SPSS. See ?read_spss or ?read_sav for more details on their optional arguments.
2.2.4.1 Exercises
[To add SPSS import exercises!]
2.2.5 Entering data directly
In R, you can construct a datasets using as.data.frame() or data.frame(). In any case, we can put small column blocks together to form a dataset as shown below.
name <- c("John Doe", "John Smith", "James Bond", "Jason Borne")
age <- runif(4, min = 35, max = 50) # this uses random prob distribution to generate age
ex_data <- data.frame(name, age)
ex_data## name age
## 1 John Doe 37.86469
## 2 John Smith 47.96433
## 3 James Bond 42.78676
## 4 Jason Borne 40.340032.3 Saving data files
2.3.1 R Data files
2.3.2 CSV spreadsheets
2.3.3 Excel spreadsheets
2.3.4 Stata datasets
2.3.5 SPSS datasets
2.3.6 Other types
delimited types entering data directly