Process Tracing and the Problem of Missing Data

被引:10
|
作者
Gonzalez-Ocantos, Ezequiel [1 ]
LaPorte, Jody [2 ]
机构
[1] Univ Oxford, Dept Polit & IR, Oxford, England
[2] Univ Oxford, Lincoln Coll, Polit & Int Relat, Oxford, England
关键词
process tracing; missing data; causal inference; causal mechanisms; case studies; CAUSAL MECHANISMS; STANDARD;
D O I
10.1177/0049124119826153
中图分类号
O1 [数学]; C [社会科学总论];
学科分类号
03 ; 0303 ; 0701 ; 070101 ;
摘要
Scholars who conduct process tracing often face the problem of missing data. The inability to document key steps in their causal chains makes it difficult to validate theoretical models. In this article, we conceptualize "missingness" as it relates to process tracing, describe different scenarios in which it is pervasive, and present three ways of addressing the problem. First, researchers should contextualize the data generation process. This requires characterizing the process whereby the actors that populate models decide whether to leave traces of their actions and motives. Researchers can thus assess whether or not incentives to produce missingness are compatible with the microfoundations of the theory, and consequently, whether or not missingness is disconfirmatory. Second, researchers may invest in indirect tests of causal mechanisms. Generating out-of-context data about microfoundations offers a plausible window into inaccessible mechanisms. Third, specifying the analytical status of steps in the causal chain allows scholars to make up for deficiencies in evidentiary support.
引用
收藏
页码:1407 / 1435
页数:29
相关论文
共 50 条
  • [11] Data in conservation: The missing link in the process
    Suenson-Taylor, K
    Sully, D
    Orton, C
    STUDIES IN CONSERVATION, 1999, 44 (03) : 184 - 194
  • [12] Learning process models with missing data
    Bridewell, Will
    Langley, Pat
    Racunas, Steve
    Borrett, Stuart
    MACHINE LEARNING: ECML 2006, PROCEEDINGS, 2006, 4212 : 557 - 565
  • [13] Hawkes Process Inference with Missing Data
    Shelton, Christian R.
    Qin, Zhen
    Shetty, Chandini
    THIRTY-SECOND AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTIETH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / EIGHTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2018, : 6425 - 6432
  • [14] Statistical analysis with missing exposure data measured by proxy respondents: a misclassification problem within a missing-data problem
    Shardell, Michelle
    Hicks, Gregory E.
    STATISTICS IN MEDICINE, 2014, 33 (25) : 4437 - 4452
  • [15] ON THE PROBLEM OF MISSING DATA IN LINEAR-MODELS
    GOURIEROUX, C
    MONFORT, A
    REVIEW OF ECONOMIC STUDIES, 1981, 48 (04): : 579 - 586
  • [16] Missing Data Problem in the Monitoring System: A Review
    Du, Jinghan
    Hu, Minghua
    Zhang, Weining
    IEEE SENSORS JOURNAL, 2020, 20 (23) : 13984 - 13998
  • [17] MISSING DATA - A PATIENT OR CLINICAL CENTER PROBLEM
    TIERNEY, C
    HALLSTORM, A
    GILLESPIE, MJ
    CONTROLLED CLINICAL TRIALS, 1986, 7 (03): : 228 - 228
  • [18] Missing boundary data recovering for the Helmholtz problem
    Ben Fatma, Riadh
    Azaiez, Mejdi
    Ben Abda, Amel
    Gmati, Nabil
    COMPTES RENDUS MECANIQUE, 2007, 335 (12): : 787 - 792
  • [19] The missing data problem in meta-analyses
    Rief, Winfried
    Hofmann, Stefan G.
    ARCHIVES OF GENERAL PSYCHIATRY, 2008, 65 (02) : 238 - 238
  • [20] Data Missing Problem in Smart Surveillance Environment
    Peixoto, Maycon L. M.
    Souza, Igo
    Barbosa, Matheus
    Lecomte, Gabriel
    Batista, Bruno G.
    Kuehne, Bruno T.
    Leite Filho, Dionisio M.
    PROCEEDINGS 2018 INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING & SIMULATION (HPCS), 2018, : 962 - 969