$30 off During Our Annual Pro Sale. View Details »

弱いつながりは転職に有効なのか? / Weak Ties and Job Change

Sansan
July 13, 2019

弱いつながりは転職に有効なのか? / Weak Ties and Job Change

■イベント
conference, on computer science for enterprise
https://ccse.jp/2019/

■登壇概要
タイトル:弱いつながりは転職に有効なのか?

登壇者:DSOC R&Dグループ 西田貴紀

▼Sansan Builders Box
https://buildersbox.corp-sansan.com/

Sansan

July 13, 2019
Tweet

More Decks by Sansan

Other Decks in Science

Transcript



  1. Dynamic Network Signatures of Labor Flows: Evidence from a Large Business Social Network
    Takanori Nishida
    (Sansan Inc., Tokyo, Japan)
    Lav R. Varshney
    (University of Illinois at Urbana-Champaign)
    Yoshiki Ishikawa
    (Habitech, Inc., Tokyo, Japan)

    View Slide

  2. Motivation




    ( )(

    View Slide

  3. Strength of Weak Ties Hypothesis

    View Slide

  4. Bird’s-Eye View
    Static
    Dynamic
    Online Offline (Face to Face)






    View Slide





  5. View Slide

  6. Data Visualization ~Dawn of Innovation~
    0 1 2 0 2

    View Slide

  7. Data
    Eight
    #'*/2015,1"2017,12"
    )3 /%&. &.Eight21 0!(
    +-$1

    View Slide

  8. User Profile


    View Slide

  9. Longitudinal Dataset (Unbalanced Panel Data)
    1st wave
    2nd wave
    3rd wave
    Jul 2015 Jan 2016 Jul 2016
    Jan 2017
    Jul 2017
    6 months 6 months
    T
    T-1
    T-2
    T
    T-1
    T-2
    6 months
    T
    T-1
    T-2
    6 months

    View Slide

  10. Empirical Definition of Weak ties and Strong Ties

    T-2 T-1 T

    =Δ Clustering coefficient
    = # of New Bridging Ties

    Ties in same community
    Bridging Ties

    View Slide

  11. View Slide



  12. View Slide

  13. Empirical Model

    ̈
    "#$
    = &' ̈
    (#$)* /$),
    + ̈
    .#$
    /ℎ121 ̈
    345
    = 345
    − 7
    34 (within transformation),8
    34
    = 1/: ∑4<=
    > 345


    W W
    W - -
    W W 1 2T
    : ; ( ) W

    View Slide

  14. Results
    ) *-
    10
    .) *-
    T-1 T
    T-2
    (
    R-squared:0.114
    Num. obs.: 417,428
    - Standard errors in parentheses
    - ***p < 0.001, **p < 0.01, *p < 0.05
    ) ( ( 0 -

    View Slide

  15. Discussion


    -


    -

    View Slide

  16. View Slide