coincidence analysis cna
play

Coincidence Analysis (CNA): A method to identify conditions - PowerPoint PPT Presentation

Coincidence Analysis (CNA): A method to identify conditions influencing implementation Deborah Cragun, PhD, MS University of South Florida College of Public Health dcragun@health.usf.edu Alanna Kulchak Rahm, PhD, MS Geisinger Health Systems


  1. Coincidence Analysis (CNA): A method to identify conditions influencing implementation Deborah Cragun, PhD, MS University of South Florida College of Public Health dcragun@health.usf.edu Alanna Kulchak Rahm, PhD, MS Geisinger Health Systems

  2. Overview 1. Why and when to use CNA 2. What CNA can and cannot do 3. Step-by-step example (hypothetical data)

  3. Path Model

  4. Fundamental Dif ifferences Inferential Statistics: Configurational Comparative Path Modeling Methods: Coincidence Analysis • Requires large sample sizes • Small to large sample sizes • Random sampling (gold standard) • Purposive sampling • Quantitative data • Quantitative or qualitative data • Independent variable or • Combinations of one or more probability of dependent variable factors (conditions) may be (holding other variables constant) needed for an outcome

  5. Conditions Outcome Home Fire Faulty electrical AND nearby couch AND other key conditions also make a difference

  6. Fundamental Dif ifferences Inferential Statistics: Configurational Comparative Path Modeling Methods: Coincidence Analysis • Requires large sample sizes • Small to large sample sizes • Random sampling (gold standard) • Purposive sampling • Quantitative data • Quantitative or qualitative data • Independent variable or • Combination of several factors probability of dependent variable may be needed for an outcome (holding other variables constant) • Uncover multiple independent paths to the same outcome

  7. Together flame source AND nearby fuel Inability to detect OR put out are minimally necessary and sufficient to start fire fire then leads to house fire #1 Home Fire More than one cause #2

  8. CNA identifies ONLY conditions that make a difference among observed cases Presence of oxygen is necessary but not a difference maker Detect & put out Source Fuel Home Fire

  9. An it iterative process Determine Select research cases question Collect Conduct CNA data Apply theory and empirical knowledge of cases

  10. Research Question Contextual Factors Implementation Outcome Strategies Successful Implementation Select cases Data Qualitative - interview transcripts Quantitative - scale measures

  11. Determine Conduct CNA research Select cases question 1. Select conditions & design and calibrate scores Collect data 2 . Evaluate data- 4. Interpret results truth table 3. Run analysis

  12. Step 1a: Select Conditions Contextual Factors Implementation Strategies • Collaborative formed held multiple planning meetings • Information accessed from LSSN website

  13. Step 1b: Calibrate Set Membership Scores • Crisp-set (0 or 1) - presence or absence of condition • 1 =full member • 0 =non-member • Fuzzy set (value between 0 and 1) - degree of each condition • .75= mostly a member of the set • .25=mostly outside set membership • Multi-value (0, 1, 2, …) finite number of values • 0= no hospitals • 1= some hospitals • 2= all hospitals

  14. House fire=0 ( hf ) Detect & put out =1 ( D ) Fuel = 0 ( f ) Source = 0 ( s ) Home Fire House fire=1 ( HF ) Source=1 ( S ) Fuel =1 ( F ) Detect & put out =0 ( d )

  15. Step 2: Evaluate Data -Truth Table Q A N E P C I S 0 0 1 1 1 1 1 1 0 1 0 0 1 1 1 1 0 0 1 1 0 1 1 1 • Ensure diversity 1 1 0 1 1 1 1 1 0 1 1 1 0 1 1 1 • Truth table can be created in the 1 1 1 0 0 1 1 1 cna package for R 1 0 0 1 1 1 1 0 1 0 0 0 0 1 1 1 0 0 1 0 0 1 1 0 • Shows configurations (patterns) 0 0 0 1 1 0 1 1 of conditions and outcomes 1 1 0 1 0 0 1 1 1 0 0 1 0 1 0 0 1 1 0 0 0 0 1 0

  16. Configur Conditions Cases -ation Contextual factors Strategies Outcome High Negative Strong Strong Peer Collaborative Informa- Successful Hospitals quality attitude of communi- leadership pressure group holds tion from implement- (N=30) evidence key person cation engage- (P) multiple LSSN tation (S) (Q) (A) networks (N) ment (E) meetings (C) website (I) LU, UR, SU, OW, c1 0 1 1 1 1 1 1 0 NW, AR, AI GL, UG, SO, SG, c2 1 0 1 0 0 1 1 1 AG c3 0 1 1 0 1 1 1 0 GR, TG c4 1 1 0 1 1 1 1 1 UH c5 0 1 1 1 0 1 1 1 BE c6 1 1 1 0 0 1 1 1 SH c7 1 0 0 1 1 1 1 0 BL c8 0 0 0 0 1 1 1 1 TI c9 0 0 0 1 0 0 1 1 VS c1 0 0 0 1 1 0 1 1 0 FR, EU c11 1 1 0 1 0 0 1 1 JU c12 1 0 0 1 0 1 0 0 VD, NE, GE, PP c13 1 0 0 0 0 1 0 1 BS, KP, GP

  17. Step 3: Run analysis in cna package for R • Default coverage and consistency threshold = 1 (can lower) • Specify “causal ordering” (if known) Peer pressure = P • Collaborative multiple Successful meetings = C Implementation = S Communication networks = N • Information accessed Leadership engagement = E from LSSN website = I Negative attitude = A Evidence quality= Q

  18. Solutions Complex solution formulas: -------------------------- outcome solution consistency coverage Atomic solution formulas: C,S (N + a*e <-> C)*(C + E <-> S) 1 0.947 ------------------------- C,S (N + a*e <-> C)*(C + a <-> S) 1 0.947 Outcome C: C,S (N + a*e <-> C)*(N+ a +E<-> S) 1 0.947 solution consistency coverage N + a*e <-> C 1 0.947 Outcome S: solution consistency coverage C + E <-> S 1 1.000 C+ a <-> S 1 0.957 N+ a + E <-> S 1 0.957

  19. Complex solution formulas: Step 4: Interpret solutions -------------------------- outcome solution C,S (N + a*e <-> C)*(C + E <-> S) C,S (N + a*e <-> C)*(a + C <-> S) C,S (N + a*e <-> C)*(a+ N +E<-> S) Strong Communication N Model ambiguity N etworks C ollaborative & or Multiple C Negative + Meetings (C) Key A ttitude – S uccessful or Key Person presence a*e Implementation + (S) Strong absence E Leadership E ngagement

  20. Complex solution formulas: Step 4: Interpret solutions -------------------------- outcome solution C,S (N + a*e <-> C)*(C + E <-> S) C,S (N + a*e <-> C)*(C + a <-> S) C,S (N + a*e <-> C)*(N+ a +E<-> S) Strong Communication Model ambiguity N etworks C ollaborative & Multiple or Negative Meetings Key A ttitude – or S uccessful presence Key Person Implementation Strong absence Leadership E ngagement

  21. Complex solution formulas: Step 4: Interpret solutions -------------------------- outcome solution C,S (N + a*e <-> C)*(C + E <-> S) C,S (N + a*e <-> C)*(C + a <-> S) C,S (N + a*e <-> C)*(N + a +E<-> S) Strong Communication Model ambiguity N etworks C ollaborative & Multiple Negative or Meetings Key A ttitude – or S uccessful Key Person presence Implementation Strong or absence Leadership E ngagement

  22. Step 4: Interpret model fit Complex solution formulas: -------------------------- Atomic solution formulas: outcome solution consistency coverage ------------------------- C,S (N + a*e <-> C)*(C + E <-> S) 1 0.947 Outcome C: C,S (N + a*e <-> C)*(C + a <-> S) 1 0.947 solution consistency coverage C,S (N + a*e <-> C)*(N+ a +E<-> S) 1 0.947 N + a*e <-> C 1 0.947 Outcome S: solution consistency coverage C + E <-> S 1 1.000 C + a <-> S 1 0.957 N + a + E <-> S 1 0.957

  23. Determine research Select cases question 1. Select conditions & design and calibrate scores Collect data 4. Interpret 2 . Evaluate data results - truth table Thank You! 3. Run analysis Questions will be held until the end of the session

  24. Configur- Conditions Cases ation Contextual factors Strategies Outcome Negative Strong Strong Collaborative Successful Hospitals attitude of communi- leadership group holds implement- (N=30) key person cation engage- multiple tation (S) (A) networks (N) ment (E) meetings (C) LU, UR, SU, OW, c1 0 1 1 1 1 NW, AR, AI GL, UG, SO, SG, c2 0 1 0 1 1 AG c3 0 1 1 1 1 GR, TG c4 1 1 0 1 1 UH c5 1 1 1 1 1 BE c6 1 1 0 1 1 SH c7 1 0 0 1 1 BL c8 0 0 0 1 1 TI c9 0 0 1 0 1 VS c1 0 0 0 1 0 1 FR, EU c11 1 0 1 0 1 JU c12 1 0 0 0 0 VD, NE, GE, PP c13 1 0 0 0 0 BS, KP, GP

Recommend


More recommend