Ron Burt prepared this data from Coleman, Katz and Menzel's 1966 study on medical innovation. They had collected data from physicians in four towns in Illinois: Peoria, Bloomington, Quincy and Galesburg. The four cities are held in one network, with each physician's city as a node attribute. No physician is in two cities, so there are no ties between cities. to_subgraphs(ison_physicians, "city") returns the four networks. Before version 2.5.0, this dataset was a list of the four networks.

Coleman, Katz and Menzel were concerned with the impact of network ties on the physicians' adoption of a new drug, tetracycline. Data on three types of ties were collected in response to three questions:

  • advice: "When you need information or advice about questions of therapy where do you usually turn?"

  • discussion: "And who are the three or four physicians with whom you most often find yourself discussing cases or therapy in the course of an ordinary week – last week for instance?"

  • friendship: "Would you tell me the first names of your three friends whom you see most often socially?"

Additional questions and records of prescriptions provided additional information:

  • recorded date of tetracycline adoption date

  • years in practice (note that these are {messydates}-compatible dates)

  • conferences attended (those that attended "Specialty" conferences presumably also attended "General" conferences)

  • regular subscriptions to medical journals

  • freetime spent associating with doctors

  • discussions, discussions on medical matters when with other doctors socially

  • clubs, memberships in clubs with other doctors

  • friends, the number of top 3 friends that are doctors

  • time practicing in current community

  • patients load (ordinal)

  • proximity, the physical proximity to other physicians (in building/sharing office)

  • medical specialty (GP/Internist/Pediatrician/Other)

  • the city where the physician practised

The physicians were interviewed with structured interviews 15 months after the drug was released. The month of adoption comes from the prescription records of the local pharmacies, not from the interviews.

data(ison_physicians)

Format

#> -- # Medical innovation --------------------------------------------------------
#> # A multiplex, directed network of 246 nodes and 310 friendship arcs, 480
#> advice arcs, and 309 discussion arcs
#>
#> -- Nodes
#> # A tibble: 246 x 13
#>   adoption specialty   conferences journals practice community patients freetime
#>      <dbl> <chr>       <chr>          <dbl> <chr>    <chr>     <chr>       <dbl>
#> 1        1 Pediatrici~ Specialty          9 1920..1~ 20+yrs    101-150       0
#> 2       12 GP          None               5 1945..   -1yr      76-100        1
#> 3        8 Internist   General            7 1935..1~ 10-20yrs  76-100        1
#> 4        9 GP          General            6 1940..1~ 5-10yrs   51-75         0
#> 5        9 GP          General            4 1935..1~ 10-20yrs  51-75         0
#> 6       10 Internist   None               7 1930..1~ 10-20yrs  101-150       0.5
#> # i 240 more rows
#> # i 5 more variables: discussions <dbl>, friends <dbl>, clubs <dbl>,
#> #   proximity <chr>, city <chr>
#>
#> -- Ties
#> # A tibble: 1,099 x 3
#>    from    to layer
#>   <int> <int> <chr>
#> 1     1     8 friendship
#> 2     1    58 friendship
#> 3     1    87 advice
#> 4     1    90 advice
#> 5     1   110 advice
#> 6     1   112 friendship
#> # i 1,093 more rows
#>

Source

{networkdata}

References

Coleman, James, Elihu Katz, and Herbert Menzel. 1966. Medical innovation: A diffusion study. Indianapolis: The Bobbs-Merrill Company.

Coleman, James, Elihu Katz, and Herbert Menzel. 1957. "The diffusion of an innovation among physicians". Sociometry 20(4): 253-270. doi:10.2307/2785979