Is there a limit to the number of levels in R?

Question

I am wondering if there is a limitation on the number of levels for a factor?

I am trying to restructure some curriculums from Xing. The selectable industries are around 135 different ones. My code looks like that, as I mentioned there are 135 different industries in my actual code.

companyIndustryLevels <- c("","ACADEMIA", "ACCOUNTING", "AEROSPACE")
levels(samples[[1]]$Industry) <- companyIndustryLevels

The following combinations work fine and are selectable when filtering the list.

genderLevels <- c("M","F")
companySizeLevels <- c("","1","1-10","11-50","51-200","201-500","501-1000","1001-5000","5001-10000","10001+")
levels(samples[[1]]$Gender) <- genderLevels
levels(samples[[1]]$CompanySize) <- companySizeLevels

So the problem is, that when viewing the list, the industry column only shows factor with 1 level, not with 135 levels.

EDIT: I am using RStudio Version 11.1.383 and R Version 3.4.3. As you can see in the reproductable example below the other columns like "Gender", "Beschäftigungsart", "Position", "Unternehmensgroesse" also got levels. When selecting the Filter in the View Window in RStudio I am able to filter all of the columns by their levels, except the "Industrie" column.

View(structure(
  list(
    ID = 1,
    Gender = structure(1L, .Label = c("M",
                                      "F"), class = "factor"),
    Bildungseinrichtungen = structure(1L, .Label = "", class = "factor"),
    Abschluss = structure(1L, .Label = "", class = "factor"),
    Studienfach = structure(1L, .Label = "", class = "factor"),
    Beschäftigungsart = structure(
      1L,
      .Label = c(
        "",
        "FULL_TIME_EMPLOYEE",
        "PART_TIME_EMPLOYEE",
        "INTERN",
        "FREELANCER",
        "OWNER",
        "PARTNER",
        "BOARD_MEMBER",
        "VOLUNTEER"
      ),
      class = "factor"
    ),
    Station.Start = NA,
    Station.Ende = NA,
    Bezeichnung = NA,
    Position = structure(
      1L,
      .Label = c(
        "",
        "STUDENT_INTERN",
        "ENTRY_LEVEL",
        "PROFESSIONAL_EXPERIENCED",
        "MANAGER_SUPERVISOR",
        "EXECUTIVE",
        "SENIOR_EXECUTIVE"
      ),
      class = "factor"
    ),
    Unternehmen = structure(1L, .Label = "AMA", class = "factor"),
    Unternehmensgroesse = structure(
      1L,
      .Label = c(
        "",
        "1",
        "1-10",
        "11-50",
        "51-200",
        "201-500",
        "501-1000",
        "1001-5000",
        "5001-10000",
        "10001+"
      ),
      class = "factor"
    ),
    Industrie = structure(
      1L,
      .Label = c(
        "ACADEMIA",
        "ACCOUNTING",
        "AEROSPACE",
        "AGRICULTURE",
        "AIRLINES",
        "ALTERNATIVE_MEDICINE",
        "APPAREL_AND_FASHION",
        "ARCHITECTURE_AND_PLANNING",
        "ARTS_AND_CRAFTS",
        "AUTOMOTIVE",
        "BANKING",
        "BIOTECHNOLOGY",
        "BROADCAST_MEDIA",
        "BUILDING_MATERIALS",
        "BUSINESS_SUPPLIES_AND_EQUIPMENT",
        "CHEMICALS",
        "CIVIC_AND_SOCIAL_ORGANIZATIONS",
        "CIVIL_ENGINEERING",
        "CIVIL_SERVICE",
        "COMPOSITES",
        "COMPUTER_AND_NETWORK_SECURITY",
        "COMPUTER_GAMES",
        "COMPUTER_HARDWARE",
        "COMPUTER_NETWORKING",
        "COMPUTER_SOFTWARE",
        "CONSTRUCTION",
        "CONSULTING",
        "CONSUMER_ELECTRONICS",
        "CONSUMER_GOODS",
        "CONSUMER_SERVICES",
        "COSMETICS",
        "DAYCARE",
        "DEFENSE_MILITARY",
        "DESIGN",
        "EDUCATION",
        "ELEARNING",
        "ELECTRICAL_ENGINEERING",
        "ENERGY",
        "ENTERTAINMENT",
        "ENVIRONMENTAL_SERVICES",
        "EVENTS_SERVICES",
        "FACILITIES_SERVICES",
        "FACILITY_MANAGEMENT",
        "FINANCIAL_SERVICES",
        "FISHERY",
        "FOOD",
        "FUNDRAISING",
        "FURNITURE",
        "GARDENING_LANDSCAPING",
        "GEOLOGY",
        "GLASS_AND_CERAMICS",
        "GRAPHIC_DESIGN",
        "HEALTH_AND_FITNESS",
        "HOSPITALITY",
        "HUMAN_RESOURCES",
        "IMPORT_AND_EXPORT",
        "INDUSTRIAL_AUTOMATION",
        "INFORMATION_SERVICES",
        "INFORMATION_TECHNOLOGY_AND_SERVICES",
        "INSURANCE",
        "INTERNATIONAL_AFFAIRS",
        "INTERNATIONAL_TRADE_AND_DEVELOPMENT",
        "INTERNET",
        "INVESTMENT_BANKING",
        "JOURNALISM",
        "LEGAL_SERVICES",
        "LEISURE_TRAVEL_AND_TOURISM",
        "LIBRARIES",
        "LOGISTICS_AND_SUPPLY_CHAIN",
        "LUXURY_GOODS_AND_JEWELRY",
        "MACHINERY",
        "MANAGEMENT_CONSULTING",
        "MARITIME",
        "MARKETING_AND_ADVERTISING",
        "MARKET_RESEARCH",
        "MECHANICAL_INDUSTRIAL_ENGINEERING",
        "MEDIA_PRODUCTION",
        "MEDICAL_DEVICES",
        "MEDICAL_SERVICES",
        "MEDICINAL_PRODUCTS",
        "METAL_METALWORKING",
        "METROLOGY_CONTROL_ENGINEERING",
        "MINING_AND_METALS",
        "MOTION_PICTURES",
        "MUSEUMS_AND_CULTURAL_INSTITUTIONS",
        "MUSIC",
        "NANOTECHNOLOGY",
        "NON_PROFIT_ORGANIZATION",
        "NURSING_AND_PERSONAL_CARE",
        "OIL_AND_ENERGY",
        "ONLINE_MEDIA",
        "OTHERS",
        "OUTSOURCING_OFFSHORING",
        "PACKAGING_AND_CONTAINERS",
        "PAPER_AND_FOREST_PRODUCTS",
        "PHOTOGRAPHY",
        "PLASTICS",
        "POLITICS",
        "PRINTING",
        "PRINT_MEDIA",
        "PROCESS_MANAGEMENT",
        "PROFESSIONAL_TRAINING_AND_COACHING",
        "PSYCHOLOGY_PSYCHOTHERAPY",
        "PUBLIC_HEALTH",
        "PUBLIC_RELATIONS_AND_COMMUNICATIONS",
        "PUBLISHING",
        "RAILROAD",
        "REAL_ESTATE",
        "RECREATIONAL_FACILITIES_AND_SERVICES",
        "RECYCLING_AND_WASTE_MANAGEMENT",
        "RENEWABLES_AND_ENVIRONMENT",
        "RESEARCH",
        "RESTAURANTS_AND_FOOD_SERVICE",
        "RETAIL",
        "SECURITY_AND_INVESTIGATIONS",
        "SEMICONDUCTORS",
        "SHIPBUILDING",
        "SPORTS",
        "STAFFING_AND_RECRUITING",
        "TAX_ACCOUNTANCY_AUDITING",
        "TELECOMMUNICATION",
        "TEXTILES",
        "THEATER_STAGE_CINEMA",
        "TIMBER",
        "TRAFFIC_ENGINEERING",
        "TRANSLATION_AND_LOCALIZATION",
        "TRANSPORT",
        "VENTURE_CAPITAL_AND_PRIVATE_EQUITY",
        "VETERINARY",
        "WELFARE_AND_COMMUNITY_HEALTH",
        "WHOLESALE",
        "WINE_AND_SPIRITS",
        "WRITING_AND_EDITING",
        "PHARMACEUTICALS"
      ),
      class = "factor"
    )
  ),
  .Names = c(
    "ID",
    "Gender",
    "Bildungseinrichtungen",
    "Abschluss",
    "Studienfach",
    "Beschäftigungsart",
    "Station.Start",
    "Station.Ende",
    "Bezeichnung",
    "Position",
    "Unternehmen",
    "Unternehmensgroesse",
    "Industrie"
  ),
  row.names = 1L,
  class = "data.frame"
))

when viewing the list the industrie column is a factor with 1 level, not 135 levels — atix, Dec 19 '17 at 12:07
Provide [reproducible](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) example: `dput(head(samples))` or `str(samples)` — zx8754, Dec 19 '17 at 12:12
Please read [How to Create a Minimal, Complete, and Verifiable Example](https://stackoverflow.com/help/mcve) and update your original post. — Len Greski, Dec 19 '17 at 12:17
To respond to your immediate question: no, there is not a limit to the number of factor levels. (This isn't absolutely true of course. I wouldn't be surprised if 2^32 is an upper bound.) There is certainly no restriction that would prevent 135 different levels. — John Coleman, Dec 19 '17 at 12:36
To follow up on my last comment, I can easily create factors with millions of levels. An attempt at a billion levels triggered a memory allocation error on my machine, but someone with more RAM might be able to pull it off. Needless to say, such RAM-stretching factors are well beyond the 135 levels you are trying to achieve. — John Coleman, Dec 19 '17 at 13:02
Eventually it is a Problem of `RStudio` especially of the *View Window in RStudio*. In this case it is not a problem of `R`. (Please correct the tags!) — jogo, Dec 19 '17 at 15:03

score 2 · Accepted Answer · answered Dec 19 '17 at 17:09

2

It seems as if the Filtering option in RStudio's Data Viewer (View()) offers a drop down menu for a factor, when its number of levels (nlevels()) is less than 65. Otherwise it defaults to a search field:

df <- data.frame(x=as.factor(1:64))
View(df)
# "filter" yields a drop down menu 

df <- data.frame(x=as.factor(1:65))
View(df)
# "filter" yields a search field 

RStudio.Version()$version
# [1] ‘1.0.143’

Note that this has nothing to do with R itself, as already mentioned in the comments.

answered Dec 19 '17 at 17:09

lukeA

53,097
5
97
100

Thank you! Then I'll go with the search field, or is there a way to get the drop down menu for more than 64 levels? – atix Dec 20 '17 at 10:37
@atix I dunno if that's possible with RStudio'S data viewer. However, if your data set is not too big, maybe use `DT::datatable(df, filter = "top")` as an alternative? – lukeA Dec 20 '17 at 18:14

Is there a limit to the number of levels in R?

1 Answers1