Convert Text in Table Variables to Categorical

This example shows how to convert a variable in a table from a cell array of character vectors to a categorical array.

Load Sample Data and Create a Table

Load sample data gathered from 100 patients.

load patients

whos
  Name                            Size            Bytes  Class      Attributes

  Age                           100x1               800  double               
  Diastolic                     100x1               800  double               
  Gender                        100x1             12212  cell                 
  Height                        100x1               800  double               
  LastName                      100x1             12416  cell                 
  Location                      100x1             15008  cell                 
  SelfAssessedHealthStatus      100x1             12340  cell                 
  Smoker                        100x1               100  logical              
  Systolic                      100x1               800  double               
  Weight                        100x1               800  double               

Store the patient data from Age, Gender, Height, Weight, SelfAssessedHealthStatus, and Location in a table. Use the unique identifiers in the variable LastName as row names.

T = table(Age,Gender,Height,Weight,...
    SelfAssessedHealthStatus,Location,...
    'RowNames',LastName);

Convert Table Variables from Cell Arrays of Character Vectors to Categorical Arrays

The cell arrays of character vectors, Gender and Location, contain discrete sets of unique values.

Convert Gender and Location to categorical arrays.

T.Gender = categorical(T.Gender);
T.Location = categorical(T.Location);

The variable, SelfAssessedHealthStatus, contains four unique values: Excellent, Fair, Good, and Poor.

Convert SelfAssessedHealthStatus to an ordinal categorical array, such that the categories have the mathematical ordering Poor < Fair < Good < Excellent.

T.SelfAssessedHealthStatus = categorical(T.SelfAssessedHealthStatus,...
    {'Poor','Fair','Good','Excellent'},'Ordinal',true);

Print a Summary

View the data type, description, units, and other descriptive statistics for each variable by using summary to summarize the table.

format compact

summary(T)
Variables:
    Age: 100x1 double
        Values:
            Min        25  
            Median     39  
            Max        50  
    Gender: 100x1 categorical
        Values:
            Female       53   
            Male         47   
    Height: 100x1 double
        Values:
            Min          60   
            Median       67   
            Max          72   
    Weight: 100x1 double
        Values:
            Min          111  
            Median     142.5  
            Max          202  
    SelfAssessedHealthStatus: 100x1 ordinal categorical
        Values:
            Poor                     11            
            Fair                     15            
            Good                     40            
            Excellent                34            
    Location: 100x1 categorical
        Values:
            County General Hospital          39    
            St. Mary s Medical Center        24    
            VA Hospital                      37    

The table variables Gender, SelfAssessedHealthStatus, and Location are categorical arrays. The summary contains the counts of the number of elements in each category. For example, the summary indicates that 53 of the 100 patients are female and 47 are male.

Select Data Based on Categories

Create a subtable, T1, containing the age, height, and weight of all female patients who were observed at County General Hospital. You can easily create a logical vector based on the values in the categorical arrays Gender and Location.

rows = T.Location=='County General Hospital' & T.Gender=='Female';

rows is a 100-by-1 logical vector with logical true (1) for the table rows where the gender is female and the location is County General Hospital.

Define the subset of variables.

vars = {'Age','Height','Weight'};

Use parentheses to create the subtable, T1.

T1 = T(rows,vars)
T1=19×3 table
                  Age    Height    Weight
                  ___    ______    ______
    Brown         49       64       119  
    Taylor        31       66       132  
    Anderson      45       68       128  
    Lee           44       66       146  
    Walker        28       65       123  
    Young         25       63       114  
    Campbell      37       65       135  
    Evans         39       62       121  
    Morris        43       64       135  
    Rivera        29       63       130  
    Richardson    30       67       141  
    Cox           28       66       111  
    Torres        45       70       137  
    Peterson      32       60       136  
    Ramirez       48       64       137  
    Bennett       35       64       131        ⋮

A is a 19-by-3 table.

Since ordinal categorical arrays have a mathematical ordering for their categories, you can perform element-wise comparisons of them with relational operations, such as greater than and less than.

Create a subtable, T2, of the gender, age, height, and weight of all patients who assessed their health status as poor or fair.

First, define the subset of rows to include in table T2.

rows = T.SelfAssessedHealthStatus<='Fair';

Then, define the subset of variables to include in table T2.

vars = {'Gender','Age','Height','Weight'};

Use parentheses to create the subtable T2.

T2 = T(rows,vars)
T2=26×4 table
                 Gender    Age    Height    Weight
                 ______    ___    ______    ______
    Johnson      Male      43       69       163  
    Jones        Female    40       67       133  
    Thomas       Female    42       66       137  
    Jackson      Male      25       71       174  
    Garcia       Female    27       69       131  
    Rodriguez    Female    39       64       117  
    Lewis        Female    41       62       137  
    Lee          Female    44       66       146  
    Hall         Male      25       70       189  
    Hernandez    Male      36       68       166  
    Lopez        Female    40       66       137  
    Gonzalez     Female    35       66       118  
    Mitchell     Male      39       71       164  
    Campbell     Female    37       65       135  
    Parker       Male      30       68       182  
    Stewart      Male      49       68       170        ⋮

T2 is a 26-by-4 table.

Related Examples

More About