Causality and Treatment Effects

Size: px
Start display at page:

Download "Causality and Treatment Effects"

Transcription

1 Causality and Treatment Effects Prof. Jacob M. Montgomery Quantitative Political Methodology (L32 363) October 24, 2016 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

2 Causal inference Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

3 Overview Thinking about causality Average treatment effects Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

4 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

5 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Why? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

6 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Why? This is a powerful, powerful way to do research Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

7 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Why? This is a powerful, powerful way to do research Building block for more advanced models Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

8 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Why? This is a powerful, powerful way to do research Building block for more advanced models Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

9 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Why? This is a powerful, powerful way to do research Building block for more advanced models NOTE: In many situations (e.g., civil war), we don t have experimental data. This means we need to think about all relevant predictors and control variables. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

10 We do not live in a univariate world So far, we ve done univariate inference. Moving forward we will do bivariate inference. This is useful: With the right study design (e.g., experimental research) Why? This is a powerful, powerful way to do research Building block for more advanced models NOTE: In many situations (e.g., civil war), we don t have experimental data. This means we need to think about all relevant predictors and control variables. We will finish the semester by talking about multiple regression as one method for attempting to deal with this. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

11 Causality In political science we want to make causal claims. X Y What does this mean? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

12 Causality In political science we want to make causal claims. X Y What does this mean? Let s do this a bit more formally for the case of an experiment (the easiest way to think about it). Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

13 Causal inference We will use T to represent a treatment variable. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

14 Causal inference We will use T to represent a treatment variable. For a categorical treatment T i = 1 if unit i receives the treatment 0 if unit i receives the control Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

15 Causal inference We will use T to represent a treatment variable. For a categorical treatment T i = 1 if unit i receives the treatment 0 if unit i receives the control We let yi 1 represent the outcome of the ith unit if the treatment is given. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

16 Causal inference We will use T to represent a treatment variable. For a categorical treatment T i = 1 if unit i receives the treatment 0 if unit i receives the control We let yi 1 represent the outcome of the ith unit if the treatment is given. We let y 0 i represent the outcome of the ith unit if the control is given. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

17 Causal inference We will use T to represent a treatment variable. For a categorical treatment T i = 1 if unit i receives the treatment 0 if unit i receives the control We let yi 1 represent the outcome of the ith unit if the treatment is given. We let yi 0 represent the outcome of the ith unit if the control is given. One of these is observed, the other is the counterfactual what would have been observed if the other treatment have been given? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

18 Causal inference We will use T to represent a treatment variable. For a categorical treatment T i = 1 if unit i receives the treatment 0 if unit i receives the control We let yi 1 represent the outcome of the ith unit if the treatment is given. We let yi 0 represent the outcome of the ith unit if the control is given. One of these is observed, the other is the counterfactual what would have been observed if the other treatment have been given? The causal effect of T i will then be yi 1 yi 0 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

19 Causal inference We will use T to represent a treatment variable. For a categorical treatment T i = 1 if unit i receives the treatment 0 if unit i receives the control We let yi 1 represent the outcome of the ith unit if the treatment is given. We let yi 0 represent the outcome of the ith unit if the control is given. One of these is observed, the other is the counterfactual what would have been observed if the other treatment have been given? The causal effect of T i will then be yi 1 yi 0 Ex., My theory is that individuals who watched this TV ad will be more likely to vote for Ted Cruz than if they didn t watch it. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

20 Average treatment effects We cannot measure individual level causal effects Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

21 Average treatment effects We cannot measure individual level causal effects We can estimate the population average treatment effect by looking at those who received the treatment and those who did not. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

22 Average treatment effects We cannot measure individual level causal effects We can estimate the population average treatment effect by looking at those who received the treatment and those who did not. ATE = mean(yi 1 yi 0) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

23 Average treatment effects We cannot measure individual level causal effects We can estimate the population average treatment effect by looking at those who received the treatment and those who did not. ATE = mean(yi 1 yi 0) ATE = mean(yi 1) mean(y i 0) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

24 Average treatment effects We cannot measure individual level causal effects We can estimate the population average treatment effect by looking at those who received the treatment and those who did not. ATE = mean(yi 1 yi 0) ATE = mean(yi 1) mean(y i 0) Each group acts as a counterfactual for the other Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

25 Average treatment effects We cannot measure individual level causal effects We can estimate the population average treatment effect by looking at those who received the treatment and those who did not. ATE = mean(yi 1 yi 0) ATE = mean(yi 1) mean(y i 0) Each group acts as a counterfactual for the other Ex., My theory is that those individuals who watched this TV ad will be more likely to vote for Mitt Romney on average than those who didn t watch it. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

26 The fundamental problem of causal inference The fundamental problem of causal inference is that at most one of yi 0 and yi 1 can be observed. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

27 The fundamental problem of causal inference The fundamental problem of causal inference is that at most one of yi 0 and yi 1 can be observed. We can think of each of these as potential outcomes. However, we can only observe one. The other is the counterfactual. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

28 The fundamental problem of causal inference The fundamental problem of causal inference is that at most one of yi 0 and yi 1 can be observed. We can think of each of these as potential outcomes. However, we can only observe one. The other is the counterfactual. Estimation of causal effects requires some combination of: Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

29 The fundamental problem of causal inference The fundamental problem of causal inference is that at most one of yi 0 and yi 1 can be observed. We can think of each of these as potential outcomes. However, we can only observe one. The other is the counterfactual. Estimation of causal effects requires some combination of: certain research designs that approximate potential outcomes Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

30 The fundamental problem of causal inference The fundamental problem of causal inference is that at most one of yi 0 and yi 1 can be observed. We can think of each of these as potential outcomes. However, we can only observe one. The other is the counterfactual. Estimation of causal effects requires some combination of: certain research designs that approximate potential outcomes randomization Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

31 The fundamental problem of causal inference The fundamental problem of causal inference is that at most one of yi 0 and yi 1 can be observed. We can think of each of these as potential outcomes. However, we can only observe one. The other is the counterfactual. Estimation of causal effects requires some combination of: certain research designs that approximate potential outcomes randomization statistical adjustment Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

32 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

33 Stand up if your student ID ends in: Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

34 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

35 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

36 Take away Causal effects rely on unobserved counterfactuals Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

37 Take away Causal effects rely on unobserved counterfactuals At best, we can estimate average treatment effects comparing those who received a treatment with those who don t. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

38 Take away Causal effects rely on unobserved counterfactuals At best, we can estimate average treatment effects comparing those who received a treatment with those who don t. In order for this to work, each group must be identical (on average) in every way except the treatment. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

39 Take away Causal effects rely on unobserved counterfactuals At best, we can estimate average treatment effects comparing those who received a treatment with those who don t. In order for this to work, each group must be identical (on average) in every way except the treatment. The best way to achieve this is through random assignment (i.e., experiments) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

40 Take away Causal effects rely on unobserved counterfactuals At best, we can estimate average treatment effects comparing those who received a treatment with those who don t. In order for this to work, each group must be identical (on average) in every way except the treatment. The best way to achieve this is through random assignment (i.e., experiments) What if this assumption is not met? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

41 Confounders and causality PROBLEM: This only works if the two groups are, on average, otherwise identical Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

42 Confounders and causality PROBLEM: This only works if the two groups are, on average, otherwise identical If the two groups differ on other factors that also cause yi 1 and y 0 this is a confounding relationship. i, Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

43 Confounders and causality PROBLEM: This only works if the two groups are, on average, otherwise identical If the two groups differ on other factors that also cause yi 1 and y 0 this is a confounding relationship. If this is the case, our counterfactual is wrong and we can make no causal claim. i, Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

44 Confounders and causality PROBLEM: This only works if the two groups are, on average, otherwise identical If the two groups differ on other factors that also cause yi 1 and y 0 this is a confounding relationship. If this is the case, our counterfactual is wrong and we can make no causal claim. i, Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

45 Confounders and causality PROBLEM: This only works if the two groups are, on average, otherwise identical If the two groups differ on other factors that also cause yi 1 and y 0 this is a confounding relationship. If this is the case, our counterfactual is wrong and we can make no causal claim. Take Away If you aren t controlling for all other relevant variables (through randomization or statistical methods), you cannot make a valid causal claim. i, Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

46 Thinking about confounding variables Direct causal relationships: X 1 Y Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

47 Thinking about confounding variables Direct causal relationships: Spurious relationships: X 1 Y Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

48 Thinking about confounding variables Direct causal relationships: Spurious relationships: X 1 Y X 2 X 1 AND X 2 Y 2 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

49 Thinking about confounding variables Direct causal relationships: Spurious relationships: X 1 Y Chain relationships: X 2 X 1 AND X 2 Y 2 Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

50 Thinking about confounding variables Direct causal relationships: Spurious relationships: X 1 Y Chain relationships: X 2 X 1 AND X 2 Y 2 X 1 X 2 Y Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

51 Thinking about confounding variables Direct causal relationships: Spurious relationships: X 1 Y Chain relationships: Multiple causation: X 2 X 1 AND X 2 Y 2 X 1 X 2 Y X 1 Y AND X 2 Y Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

52 Thinking about confounding variables Direct causal relationships: Spurious relationships: X 1 Y Chain relationships: Multiple causation: X 2 X 1 AND X 2 Y 2 X 1 X 2 Y X 1 Y AND X 2 Y Direct and indirect causation: X 1 Y AND X 1 X 2 AND X 2 Y Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

53 Write down one of each type of claim for this data. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

54 Being a responsible causal analyst Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

55 Less silly examples Vitamin C? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

56 Less silly examples Vitamin C? Flossing? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

57 Less silly examples Vitamin C? Flossing? Drinking while pregnant? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

58 Less silly examples Vitamin C? Flossing? Drinking while pregnant? Fox News? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

59 At a minimum we need to show... Association What we will be doing this rest of the semester Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

60 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

61 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Association = causation Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

62 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Association = causation Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

63 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Association = causation Temporal order For T i to cause Y i it must come before Y in time order Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

64 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Association = causation Temporal order For T i to cause Y i it must come before Y in time order Post hoc ergo propter hoc Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

65 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Association = causation Temporal order For T i to cause Y i it must come before Y in time order Post hoc ergo propter hoc After this, therefore because of this Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

66 At a minimum we need to show... Association What we will be doing this rest of the semester Correlation, contingency tables, regression coefficients,... Association = causation Temporal order For T i to cause Y i it must come before Y in time order Post hoc ergo propter hoc After this, therefore because of this Temporal order does = causation (e.g., every superstition ever) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

67 At a minimum we need to show... Eliminate alternative explanations Suppose there is an association and a proper time order. We stil cannot infer causation. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

68 At a minimum we need to show... Eliminate alternative explanations Suppose there is an association and a proper time order. We stil cannot infer causation. Rather, we must test for all alternative explanations. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

69 At a minimum we need to show... Eliminate alternative explanations Suppose there is an association and a proper time order. We stil cannot infer causation. Rather, we must test for all alternative explanations. Only if all of these have been resolved can we claim causation. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

70 At a minimum we need to show... Eliminate alternative explanations Suppose there is an association and a proper time order. We stil cannot infer causation. Rather, we must test for all alternative explanations. Only if all of these have been resolved can we claim causation. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

71 At a minimum we need to show... Eliminate alternative explanations Suppose there is an association and a proper time order. We stil cannot infer causation. Rather, we must test for all alternative explanations. Only if all of these have been resolved can we claim causation. How can we do this? Experimental control Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

72 At a minimum we need to show... Eliminate alternative explanations Suppose there is an association and a proper time order. We stil cannot infer causation. Rather, we must test for all alternative explanations. Only if all of these have been resolved can we claim causation. How can we do this? Experimental control Statistical control (Stay tuned...) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

73 Problem: It can be subtle (Mount 2010) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

74 A case study Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

75 Problem: It can be subtle Rogers, Coffman, and Bergman: While the authors control for certain variables, their research only implies there is a relationship between parental involvement and student performance. This caveat is important; the existence of a relationship does not tell us what causes what. Think of it this way: If you had two children, and one was getting A s and the other C s, which of them would you help more? The C student. An outsider, noticing that you ve spent the school year helping only one of your children, might infer that parental help caused that child to earn lower grades. This of course would not be the case, and inferring causation here would be a mistake. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

76 If you see a surprising result, be skeptical Occasional Notes 35 Nobel Laureates per 10 Million Population r=0.791 P< The Netherlands Poland Portugal Greece Japan China Brazil Belgium Canada Italy Spain United States France Australia Sweden Austria Denmark Finland Ireland Norway United Kingdom Switzerland Germany Chocolate Consumption (kg/yr/capita) Figure 1. Correlation between Countries Annual Per Capita Chocolate Consumption and the Number of Nobel Laureates per 10 Million Population. Discussion about 14 Nobel laureates, yet we observe 32. Lecture 13 (QPM 2016) Causality and Treatment Considering Effects that in this instance the October observed 24, / 32

77 Example: Can social pressure increase turnout? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

78 Comparing two populations We have two independent samples, and we want to compare them. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

79 Comparing two populations We have two independent samples, and we want to compare them. Our data will look like this. Variable 1 Variable 2 (Outcome or response) (Explanatory or grouping) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

80 Example: Social Pressure and Turnout Gerber, A., Green, D., and Larimer C.W Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment. American Political Science Review 101(1): Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

81 Example: Social Pressure and Turnout Gerber, A., Green, D., and Larimer C.W Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment. American Political Science Review 101(1): Obviously, most of these treatments worked. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

82 Example: Social Pressure and Turnout Gerber, A., Green, D., and Larimer C.W Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment. American Political Science Review 101(1): Obviously, most of these treatments worked. (Note: The analysis of this experiment is actually somewhat more complicated than this due to the way individuals were assigned to treatment groups.) Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

83 Do Politicians Racially Discriminate? Broad Question Is racial discrimination against blacks still a problem in the political sphere? Puzzle Do legislators discriminate against individual requests for constituency service on the basis of race? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

84 Tuesday, October 25, 2011 Tuesday, October 25, 2011 Who do you think they are? Jake Mueller DeShawn Jackson Jake Mueller Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

85 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white, Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

86 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

87 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

88 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

89 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Experiment Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

90 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Experiment The sample includes states legislators in 44 U.S. states with a valid address in September Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

91 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Experiment The sample includes states legislators in 44 U.S. states with a valid address in September Race was signaled by randomizing whether the was signed and sent from an account with the name Jake Mueller or DeShawn Jackson. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

92 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Experiment The sample includes states legislators in 44 U.S. states with a valid address in September Race was signaled by randomizing whether the was signed and sent from an account with the name Jake Mueller or DeShawn Jackson. The text of the was also manipulated so as to signal the partisan preference of the sender. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

93 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Experiment The sample includes states legislators in 44 U.S. states with a valid address in September Race was signaled by randomizing whether the was signed and sent from an account with the name Jake Mueller or DeShawn Jackson. The text of the was also manipulated so as to signal the partisan preference of the sender. The cross-tabulation between race & partisan preference gives six treatments (or groups). Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

94 Comparing legislators responsiveness Questions: Do legislators answer a higher proportion of s from the citizens they believe are white,... even though both black & white don t signal party affiliation?... even though both black & white signal being Republican?... even though both black & white signal being Democrat? Experiment The sample includes states legislators in 44 U.S. states with a valid address in September Race was signaled by randomizing whether the was signed and sent from an account with the name Jake Mueller or DeShawn Jackson. The text of the was also manipulated so as to signal the partisan preference of the sender. The cross-tabulation between race & partisan preference gives six treatments (or groups). The outcome variable is the response (or lack thereof) to any s. Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

95 Results: Table 1?? Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

96 Class business Midterms PS posted today. Quiz for next class over t-tests Lecture 13 (QPM 2016) Causality and Treatment Effects October 24, / 32

Multivariate relationships. Week 6 22 February, 2016 Prof. Andrew Eggers

Multivariate relationships. Week 6 22 February, 2016 Prof. Andrew Eggers Multivariate relationships Week 6 22 February, 2016 Prof. Andrew Eggers 1 2 The new england journal of medicine occasional notes Chocolate Consumption, Cognitive Function, and Nobel Laureates Franz H.

More information

STATS Relationships between variables: Correlation

STATS Relationships between variables: Correlation STATS 1060 Relationships between variables: Correlation READINGS: Chapter 7 of your text book (DeVeaux, Vellman and Bock); on-line notes for correlation; on-line practice problems for correlation NOTICE:

More information

WESTERN EUROPE PREVALENCE AND INCIDENCE OF PERIPHERAL ARTERY DISEASE AND CRITICAL LIMB ISCHEMIA 2017

WESTERN EUROPE PREVALENCE AND INCIDENCE OF PERIPHERAL ARTERY DISEASE AND CRITICAL LIMB ISCHEMIA 2017 WESTERN EUROPE PREVALENCE AND INCIDENCE OF PERIPHERAL ARTERY DISEASE AND CRITICAL LIMB ISCHEMIA 2017 Mary L. Yost 404-520-6652 , LLC 23 Ridge Rd. Beaufort, SC 20097 Copyright Pending 2017 All rights reserved,

More information

Political Science 15, Winter 2014 Final Review

Political Science 15, Winter 2014 Final Review Political Science 15, Winter 2014 Final Review The major topics covered in class are listed below. You should also take a look at the readings listed on the class website. Studying Politics Scientifically

More information

Case study examining the impact of German reunification on life expectancy

Case study examining the impact of German reunification on life expectancy Supplementary Materials 2 Case study examining the impact of German reunification on life expectancy Table A1 summarises our case study. This is a simplified analysis for illustration only and does not

More information

Global EHS Resource Center

Global EHS Resource Center Global EHS Resource Center Understand environmental and workplace safety requirements that affect your global operations. 800.372.1033 bna.com/gelw Global EHS Resource Center This comprehensive research

More information

Chapter Eight: Multivariate Analysis

Chapter Eight: Multivariate Analysis Chapter Eight: Multivariate Analysis Up until now, we have covered univariate ( one variable ) analysis and bivariate ( two variables ) analysis. We can also measure the simultaneous effects of two or

More information

Chapter Eight: Multivariate Analysis

Chapter Eight: Multivariate Analysis Chapter Eight: Multivariate Analysis Up until now, we have covered univariate ( one variable ) analysis and bivariate ( two variables ) analysis. We can also measure the simultaneous effects of two or

More information

The influence of societal individualism on a century of tobacco use: modelling the prevalence of smoking Appendices A and B

The influence of societal individualism on a century of tobacco use: modelling the prevalence of smoking Appendices A and B The influence of societal individualism on a century of tobacco use: modelling the prevalence of smoking Appendices A and B J.C. Lang, D.M. Abrams, and H. De Sterck A Additional Tables and Figures Table

More information

Chapter 11 Nonexperimental Quantitative Research Steps in Nonexperimental Research

Chapter 11 Nonexperimental Quantitative Research Steps in Nonexperimental Research Chapter 11 Nonexperimental Quantitative Research (Reminder: Don t forget to utilize the concept maps and study questions as you study this and the other chapters.) Nonexperimental research is needed because

More information

508 the number of suicide deaths in deaths per 100,000 people was the suicide rate in Suicide deaths in 2013 by gender

508 the number of suicide deaths in deaths per 100,000 people was the suicide rate in Suicide deaths in 2013 by gender An overview of suicide statistics This document summarises information about suicide deaths in New Zealand covering up to 13. It does not attempt to explain causes of suicidal behaviour or causes of changes

More information

Smokefree Policies in Europe: Are we there yet?

Smokefree Policies in Europe: Are we there yet? Smokefree Policies in Europe: Are we there yet? 14 April 2015, 9:00 10:30am Rue de l Industrie 24, 1040 Brussels Permanent Partners: Temporary Partners: The research for the SFP Smokefree Map was partially

More information

PRODUCT INFORMATION. New Reagents for Dako CoverStainer. Choose the H&E staining intensity you want.

PRODUCT INFORMATION. New Reagents for Dako CoverStainer. Choose the H&E staining intensity you want. PRODUCT INFORMATION Primary Staining Dako CoverStainer Reagents New Reagents for Dako CoverStainer. Choose the H&E staining intensity you want. New flexible staining intensities to fit y Dako H&E Staining

More information

Design and Analysis of a Cancer Prevention Trial: Plans and Results. Matthew Somerville 09 November 2009

Design and Analysis of a Cancer Prevention Trial: Plans and Results. Matthew Somerville 09 November 2009 Design and Analysis of a Cancer Prevention Trial: Plans and Results Matthew Somerville 09 November 2009 Overview Objective: Review the planned analyses for a large prostate cancer prevention study and

More information

Bio-Rad Laboratories HEMOGLOBIN Testing Bio-Rad A1c. VARIANT II TURBO Link. Fully-Automated HbA 1c Testing

Bio-Rad Laboratories HEMOGLOBIN Testing Bio-Rad A1c. VARIANT II TURBO Link. Fully-Automated HbA 1c Testing Bio-Rad Laboratories HEMOGLOBIN Testing Bio-Rad A1c VARIANT II TURBO Link Fully-Automated HbA 1c Testing Bio-Rad Laboratories HEMOGLOBIN Testing VARIANT II TURBO Link Efficient streamlined process Powerful

More information

A. Indicate the best answer to each the following multiple-choice questions (20 points)

A. Indicate the best answer to each the following multiple-choice questions (20 points) Phil 12 Fall 2012 Directions and Sample Questions for Final Exam Part I: Correlation A. Indicate the best answer to each the following multiple-choice questions (20 points) 1. Correlations are a) useful

More information

Module 4 Introduction

Module 4 Introduction Module 4 Introduction Recall the Big Picture: We begin a statistical investigation with a research question. The investigation proceeds with the following steps: Produce Data: Determine what to measure,

More information

Deceased donation data in the UK. Paul Murphy National Clinical Lead for Organ Donation United Kingdom

Deceased donation data in the UK. Paul Murphy National Clinical Lead for Organ Donation United Kingdom Deceased donation data in the UK Paul Murphy National Clinical Lead for Organ Donation United Kingdom Deceased donation data in the UK And the story behind it Paul Murphy National Clinical Lead for Organ

More information

Department of Paediatric Chemical Pathology and Neonatal Screening The Children s Hospital Sheffield S10 2TH UK. Annual Report 2002.

Department of Paediatric Chemical Pathology and Neonatal Screening The Children s Hospital Sheffield S10 2TH UK. Annual Report 2002. Ms M Downing Dr J R Bonham Professor R J Pollitt Telephone (+)44 114 271 7404 Fax (+)44 114 276 6205 Department of Paediatric Chemical Pathology and Neonatal Screening The Children s Hospital Sheffield

More information

HRM Series PCB Power Relays

HRM Series PCB Power Relays Features: SPCO contacts. Sealed to allow washing after flow soldering. Specifications: Coil Data: Nominal Voltage Nominal Power consumption : 3V dc to 24V dc. : 540mW to 720mW. Contact Data: Contact Arrangement

More information

Type 1 Diabetes Australian Research Impact Analysis

Type 1 Diabetes Australian Research Impact Analysis Type 1 Diabetes Australian Research Impact Analysis Executive Overview Summary Type 1 diabetes research in Australia Australia is making a significant contribution to the quantity and quality of the global

More information

WCPT COUNTRY PROFILE December 2017 SWEDEN

WCPT COUNTRY PROFILE December 2017 SWEDEN WCPT COUNTRY PROFILE December 2017 SWEDEN SWEDEN NUMBERS WCPT Members Practising physical therapists 11,043 Total number of physical therapist members in your member organisation 17,906 Total number of

More information

The OECD Health Care Quality Indicators Project

The OECD Health Care Quality Indicators Project The OECD Health Care Quality Indicators Project Ed Kelley, Ph.D. Head, OECD Health Care Quality Indicators Project Health Systems Working Party Luxembourg - April 26, 2005 1 Broad aims of the OECD s HCQI

More information

PROMOTION AND PROTECTION OF ALL HUMAN RIGHTS, CIVIL, POLITICAL, ECONOMIC, SOCIAL AND CULTURAL RIGHTS, INCLUDING THE RIGHT TO DEVELOPMENT

PROMOTION AND PROTECTION OF ALL HUMAN RIGHTS, CIVIL, POLITICAL, ECONOMIC, SOCIAL AND CULTURAL RIGHTS, INCLUDING THE RIGHT TO DEVELOPMENT UNITED NATIONS A General Assembly Distr. LIMITED A/HRC/11/L.16 16 June 2009 Original: ENGLISH HUMAN RIGHTS COUNCIL Eleventh session Agenda item 3 PROMOTION AND PROTECTION OF ALL HUMAN RIGHTS, CIVIL, POLITICAL,

More information

(CORRELATIONAL DESIGN AND COMPARATIVE DESIGN)

(CORRELATIONAL DESIGN AND COMPARATIVE DESIGN) UNIT 4 OTHER DESIGNS (CORRELATIONAL DESIGN AND COMPARATIVE DESIGN) Quasi Experimental Design Structure 4.0 Introduction 4.1 Objectives 4.2 Definition of Correlational Research Design 4.3 Types of Correlational

More information

Allied Health: Sustainable Integrated Health Care for all Australians

Allied Health: Sustainable Integrated Health Care for all Australians Allied Health: Sustainable Integrated Health Care for all Australians Catherine Turnbull Chief Allied and Scientific Health Advisor SA Health Presentation to Indigenous Allied Health Australia Conference,

More information

WCPT COUNTRY PROFILE December 2017 HUNGARY

WCPT COUNTRY PROFILE December 2017 HUNGARY WCPT COUNTRY PROFILE December 2017 HUNGARY HUNGARY NUMBERS WCPT Members Practising physical therapists 727 Total number of physical therapist members in your member organisation 4,000 Total number of practising

More information

WCPT COUNTRY PROFILE December 2017 SERBIA

WCPT COUNTRY PROFILE December 2017 SERBIA WCPT COUNTRY PROFILE December 2017 SERBIA SERBIA NUMBERS WCPT Members Practising physical therapists 622 Total number of physical therapist members in your member organisation 3,323 Total number of practising

More information

Political Science 30: Political Inquiry Section 1

Political Science 30: Political Inquiry Section 1 Political Science 30: Political Inquiry Section 1 Taylor Carlson tncarlson@ucsd.edu January 10, 2019 Carlson POLI 30-Section 1 January 10, 2019 1 / 16 Learning Outcomes By the end of section today, you

More information

Bio-Rad Laboratories HEMOGLOBIN Testing Bio-Rad A1c. VARIANT II TURBO Link System. Fully-Automated HbA 1c Testing

Bio-Rad Laboratories HEMOGLOBIN Testing Bio-Rad A1c. VARIANT II TURBO Link System. Fully-Automated HbA 1c Testing Bio-Rad Laboratories HEMOGLOBIN Testing Bio-Rad A1c VARIANT II TURBO Link System Fully-Automated HbA 1c Testing Bio-Rad Laboratories HEMOGLOBIN Testing VARIANT II TURBO Link System Efficient streamlined

More information

HPV, Cancer and the Vaccination Programme

HPV, Cancer and the Vaccination Programme HPV, Cancer and the Vaccination Programme Dr Brenda Corcoran Consultant in Public Health Medicine HSE National Immunisation Office What is HPV? Human papillomavirus Over 200 types identified Spread by

More information

by EUA Denmark Italy Portugal Israel Canada Finland Luxembourg Spain Australia Mexico France Monaco Sweden China Austria Germany Netherlands Switzerland Japan Belgium Ireland Norway United Kingdom South

More information

IOSH No Time to Lose campaign: working together to tackle asbestos-related cancer #NTTLasbestos. Jonathan Hughes IOSH Vice President

IOSH No Time to Lose campaign: working together to tackle asbestos-related cancer #NTTLasbestos. Jonathan Hughes IOSH Vice President IOSH No Time to Lose campaign: working together to tackle asbestos-related cancer #NTTLasbestos Jonathan Hughes IOSH Vice President About the Institution of Occupational Safety and Health (IOSH) www.iosh.co.uk

More information

Gender differences in competitive preferences: new cross-country empirical evidence

Gender differences in competitive preferences: new cross-country empirical evidence SCHUMPETER DISCUSSION PAPERS Gender differences in competitive preferences: new cross-country empirical evidence Werner Bönte SDP 2014-008 ISSN 1867-5352 by the author Gender differences in competitive

More information

MENTAL HEALTH CARE. OECD HCQI Expert meeting 17 th of May, Rie Fujisawa

MENTAL HEALTH CARE. OECD HCQI Expert meeting 17 th of May, Rie Fujisawa MENTAL HEALTH CARE OECD HCQI Expert meeting 17 th of May, 2013 Rie Fujisawa Mental health indicators Any hospital readmissions for patients with schizophrenia Same hospital readmissions for patients with

More information

Burden and cost of alcohol, tobacco and illegal drugs globally and in Europe

Burden and cost of alcohol, tobacco and illegal drugs globally and in Europe Burden and cost of alcohol, tobacco and illegal drugs globally and in Europe Jürgen Rehm 1-4 Kevin D. Shield 1,2,3 1) Centre for Addiction and Mental Health, Toronto, Canada 2) University of Toronto, Canada

More information

An analysis of structural changes in the consumption patterns of beer, wine and spirits applying an alternative methodology

An analysis of structural changes in the consumption patterns of beer, wine and spirits applying an alternative methodology An analysis of structural changes in the consumption patterns of beer, wine and spirits applying an alternative methodology 1. Introduction Jan BENTZEN, Valdemar SMITH Department of Economics and Business

More information

10. Introduction to Multivariate Relationships

10. Introduction to Multivariate Relationships 10. Introduction to Multivariate Relationships Bivariate analyses are informative, but we usually need to take into account many variables. Many explanatory variables have an influence on any particular

More information

Unit 7 Comparisons and Relationships

Unit 7 Comparisons and Relationships Unit 7 Comparisons and Relationships Objectives: To understand the distinction between making a comparison and describing a relationship To select appropriate graphical displays for making comparisons

More information

Men & Health Work. Difference can make a difference Steve Boorman & Ian Banks RSPH Academy 2013

Men & Health Work. Difference can make a difference Steve Boorman & Ian Banks RSPH Academy 2013 Men & Health Promotion @ Work Difference can make a difference Steve Boorman & Ian Banks RSPH Academy 2013 Difference can make a Difference Mens health: State of mens health Use of services Role of the

More information

Louisville '19 Attachment #69

Louisville '19 Attachment #69 Telephone Meeting Approved and why I propose Using zoom to fulfill both Phone and Virtual video meeting Formats. The first established phone meeting Sanctioned by Gamblers Anonymous (listed on Trustee

More information

The health economic landscape of cancer in Europe

The health economic landscape of cancer in Europe 1 Approval number The health economic landscape of cancer in Europe Bengt Jönsson, Professor Emeritus of Health Economics Stockholm School of Economics 2 Disclaimer This presentation was developed by Professor

More information

Immunohematology. IH-QC Modular System. Select. Combine. Control.

Immunohematology. IH-QC Modular System. Select. Combine. Control. Immunohematology IH-QC Modular System Select. Combine. Control. IH-QC Modular System Select. Combine. Control. Transfusion guidelines recommend regular checking of test materials, test methods, local working

More information

Chapter 4: More about Relationships between Two-Variables Review Sheet

Chapter 4: More about Relationships between Two-Variables Review Sheet Review Sheet 4. Which of the following is true? A) log(ab) = log A log B. D) log(a/b) = log A log B. B) log(a + B) = log A + log B. C) log A B = log A log B. 5. Suppose we measure a response variable Y

More information

Section 2: Data & Measurement

Section 2: Data & Measurement Section 2: Data & Measurement Michael Gill Gov 50 September 21, 2011 Outline (0) Questions/Concerns (1) Review of Lecture (2) Measurement (3) Sampling Last Week Last week, we talked about theories, which

More information

Lecture (chapter 1): Introduction

Lecture (chapter 1): Introduction Lecture (chapter 1): Introduction Ernesto F. L. Amaral January 17, 2018 Advanced Methods of Social Research (SOCI 420) Source: Healey, Joseph F. 2015. Statistics: A Tool for Social Research. Stamford:

More information

The Risk of Alcohol in Europe. Bridging the Gap June 2004

The Risk of Alcohol in Europe. Bridging the Gap June 2004 The Risk of Alcohol in Europe Bridging the Gap 16-19 June 2004 1. What is the relationship between alcohol and the risk of heart disease? 2. What is the relationship between alcohol and the risk of other

More information

DEPARTMENT OF POLITICAL SCIENCE AND INTERNATIONAL RELATIONS Research Methods Posc 302 ANALYSIS OF SURVEY DATA

DEPARTMENT OF POLITICAL SCIENCE AND INTERNATIONAL RELATIONS Research Methods Posc 302 ANALYSIS OF SURVEY DATA DEPARTMENT OF POLITICAL SCIENCE AND INTERNATIONAL RELATIONS Research Methods Posc 302 ANALYSIS OF SURVEY DATA I. TODAY S SESSION: A. Second steps in data analysis and interpretation 1. Examples and explanation

More information

PARALLELISM AND THE LEGITIMACY GAP 1. Appendix A. Country Information

PARALLELISM AND THE LEGITIMACY GAP 1. Appendix A. Country Information PARALLELISM AND THE LEGITIMACY GAP 1 Appendix A Country Information PARALLELISM AND THE LEGITIMACY GAP 2 Table A.1 Sample size by country 2006 2008 2010 Austria 2405 2255 0 Belgium 1798 1760 1704 Bulgaria

More information

King, Keohane and Verba, Designing Social Inquiry, pp. 3-9, 36-46, notes for Comparative Politics Field Seminar, Fall, 1998.

King, Keohane and Verba, Designing Social Inquiry, pp. 3-9, 36-46, notes for Comparative Politics Field Seminar, Fall, 1998. King, Keohane and Verba, Designing Social Inquiry, pp. 3-9, 36-46, 115-149 notes for Comparative Politics Field Seminar, Fall, 1998. (pp. 1-9) King, Keohane and Verba s (from now on KKV) primary goal is

More information

HER2 FISH pharmdx TM Interpretation Guide - Breast Cancer

HER2 FISH pharmdx TM Interpretation Guide - Breast Cancer P A T H O L O G Y HER2 FISH pharmdx TM Interpretation Guide - Breast Cancer For In Vitro Diagnostic Use FDA approved as an aid in the assessment of patients for whom Herceptin TM (trastuzumab) treatment

More information

QPM Lab 9: Contingency Tables and Bivariate Displays in R

QPM Lab 9: Contingency Tables and Bivariate Displays in R QPM Lab 9: Contingency Tables and Bivariate Displays in R Department of Political Science Washington University, St. Louis November 3-4, 2016 QPM Lab 9: Contingency Tables and Bivariate Displays in R 1

More information

EFSA s Concise European food consumption database. Davide Arcella Data Collection and Exposure Unit

EFSA s Concise European food consumption database. Davide Arcella Data Collection and Exposure Unit EFSA s Concise European food consumption database Davide Arcella Data Collection and Exposure Unit 1 The EFSA raison d être Risk assessment authority created in 2002 as part of a comprehensive program

More information

Quantitative Analysis and Empirical Methods

Quantitative Analysis and Empirical Methods 2) Sciences Po, Paris, CEE / LIEPP University of Gothenburg, CERGU / Political Science 1 Explaining Relationships 2 Introduction We have learned about converting concepts into variables Our key interest

More information

CARDIOVASCULAR DISEASE AND DIABETES:

CARDIOVASCULAR DISEASE AND DIABETES: CARDIOVASCULAR DISEASE AND DIABETES: HOW OECD HEALTH SYSTEMS DELIVER BETTER OUTCOMES? Progress report 7 th November 2013 Outline Overview of the project Preliminary results Descriptive Analytical Next

More information

- Network for Excellence in Health Innovation

- Network for Excellence in Health Innovation Real-world evidence is evidence from any and all sources of data that may contribute to more effective health care, including health care best tailored to the needs of individual patients. - Network for

More information

Biology Report. Is there a relationship between Countries' Human Development Index (HDI) level and the incidence of tuberculosis?

Biology Report. Is there a relationship between Countries' Human Development Index (HDI) level and the incidence of tuberculosis? Biology Report Is there a relationship between Countries' Human Development Index (HDI) level and the incidence of tuberculosis? Introduction Tuberculosis is a serious disease caused by the bacterium Mycobacterium

More information

LEBANON. WCPT COUNTRY PROFILE December 2018

LEBANON. WCPT COUNTRY PROFILE December 2018 LEBANON WCPT COUNTRY PROFILE December 2018 LEBANON NUMBERS 1600 1400 1200 1000 800 600 400 200 0 Physical therapists in the country Members in MO 1,480 1,480 Total PTs in country 800000 700000 600000 500000

More information

EUROPEAN AEROSOL PRODUCTION

EUROPEAN AEROSOL PRODUCTION Introduction EUROPEAN AEROSOL PRODUCTION 2017 EU European production Market share Production by segment Worldwide production / FEA, 2018. 1 INTRODUCTION More than 5.7 billion over 16 billion units globally

More information

GENERAL PRINCIPLES ON FLEXIBILITY OF WORDING FOR HEALTH CLAIMS

GENERAL PRINCIPLES ON FLEXIBILITY OF WORDING FOR HEALTH CLAIMS GENERAL PRINCIPLES ON FLEXIBILITY OF WORDING FOR HEALTH CLAIMS These general principles were presented for the first time at an informal meeting in Brussels on 19 June 2012. Experts from 17 Member States

More information

Arab American Voters 2014

Arab American Voters 2014 Arab American Voters 2014 Their Identity and Political Concerns November 24, 2014 Executive Summary Identity and Personal Concerns Ethnic pride and identity remains high among Arab Americans. A majority

More information

Manuel Cardoso RARHA Executive Coordinator Public Health MD Senior Advisor Deputy General-Director of SICAD - Portugal

Manuel Cardoso RARHA Executive Coordinator Public Health MD Senior Advisor Deputy General-Director of SICAD - Portugal Manuel Cardoso RARHA Executive Coordinator Public Health MD Senior Advisor Deputy General-Director of SICAD - Portugal Public Health Public health is the science and art of preventing disease, prolonging

More information

Collecting Data: Observational Studies

Collecting Data: Observational Studies STAT 250 Dr. Kari Lock Morgan Data Collection and Bias Collecting Data: Observational Studies SECTION 1.3 Association versus Causation Confounding s Observational Studies vs Experiments Population Sample

More information

University of Oxford Intermediate Social Statistics: Lecture One

University of Oxford Intermediate Social Statistics: Lecture One University of Oxford Intermediate Social Statistics: Lecture One Raymond M. Duch Nuffield College Oxford www.raymondduch.com @RayDuch January 17, 2012 Course Requirements Course Requirements Eight Lectures

More information

Request for Letters of Intent. International Development of H5N1 Influenza Vaccines

Request for Letters of Intent. International Development of H5N1 Influenza Vaccines Request for Letters of Intent International Development of H5N1 Influenza Vaccines The World Health Organization (WHO) intends to provide funding to developing (1) country vaccine manufacturers to develop

More information

BioPlex 2200 Infectious Disease Panels

BioPlex 2200 Infectious Disease Panels BioPlex 2200 System BioPlex 2200 Infectious Disease Panels An Expanding Multiplexed Assay Menu Lyme HIV Ag-Ab MMV IgM Syphilis Total & RPR MMRV EBV HSV-1 & HSV-2 EBV IgM ToRC IgM ToRC Leading the way with

More information

Drinking guidelines used in the context of early identification and brief interventions in Europe: overview of RARHA survey results

Drinking guidelines used in the context of early identification and brief interventions in Europe: overview of RARHA survey results Drinking guidelines used in the context of early identification and brief interventions in Europe: overview of RARHA survey results E. Scafato, C. Gandin, L. Galluzzo, S. Ghirini, S. Martire Istituto Superiore

More information

Hearing Loss: The Statistics

Hearing Loss: The Statistics : The Statistics 2015 Global Statistics It is hard to know precise numbers of how many people experience hearing loss across the EU, Europe, and indeed the world. There are many sources of information

More information

DENMARK. WCPT COUNTRY PROFILE December 2018

DENMARK. WCPT COUNTRY PROFILE December 2018 DENMARK WCPT COUNTRY PROFILE December 2018 DENMARK NUMBERS 14000 12000 10000 8000 6000 4000 2000 0 Physical therapists in the country Members in MO 11,720 12,975 Total PTs in country 800000 700000 600000

More information

Alcohol-related harm in Europe and the WHO policy response

Alcohol-related harm in Europe and the WHO policy response Alcohol-related harm in Europe and the WHO policy response Lars Moller Programme Manager World Health Organization Regional Office for Europe Date of presentation NCD global monitoring framework: alcohol-related

More information

Cross Border Genetic Testing for Rare Diseases

Cross Border Genetic Testing for Rare Diseases Cross Border Genetic Testing for Rare Diseases EUCERD Joint Action WP8 Helena Kääriäinen National Institute for Health an Welfare, Helsinki, Finland Starting point Possibilities and demand for genetic

More information

Results you can trust

Results you can trust PRODUCT I NF OR MAT ION pharmdx Results you can trust The first and only FDA-approved PD-L1 test to assess the magnitude of treatment effect on progression-free survival in melanoma patients from OPDIVO

More information

Yersiniosis SURVEILLANCE REPORT. Annual Epidemiological Report for Key facts. Methods. Epidemiology

Yersiniosis SURVEILLANCE REPORT. Annual Epidemiological Report for Key facts. Methods. Epidemiology SURVEILLANCE REPORT Annual Epidemiological Report for 2015 Yersiniosis Key facts In 2015, 26 countries reported 7 279 confirmed yersiniosis cases in the EU/EEA. The overall notification rate was 2.0 cases

More information

Sponsor. Generic Drug Name. Trial Indication(s) Protocol Number. Protocol Title. Clinical Trial Phase. Study Start/End Dates. Novartis.

Sponsor. Generic Drug Name. Trial Indication(s) Protocol Number. Protocol Title. Clinical Trial Phase. Study Start/End Dates. Novartis. Sponsor Novartis Generic Drug Name Lumiracoxib Trial Indication(s) Not applicable Protocol Number CCOX189A2483 Protocol Title A retrospective pharmacogenetics analysis of patients with elevated liver enzymes

More information

Thomas Karlsson & Esa Österberg National Research and Development Centre for Welfare and Health Alcohol and Drug Research Group P.O.

Thomas Karlsson & Esa Österberg National Research and Development Centre for Welfare and Health Alcohol and Drug Research Group P.O. European Comparative Alcohol Study Europe and Alcohol Policy Thomas Karlsson & Esa Österberg National Research and Development Centre for Welfare and Health Alcohol and Drug Research Group P.O.BO 220 FIN-00531

More information

Stat 13, Intro. to Statistical Methods for the Life and Health Sciences.

Stat 13, Intro. to Statistical Methods for the Life and Health Sciences. Stat 13, Intro. to Statistical Methods for the Life and Health Sciences. 0. SEs for percentages when testing and for CIs. 1. More about SEs and confidence intervals. 2. Clinton versus Obama and the Bradley

More information

World Connections Committee (WCC) Report

World Connections Committee (WCC) Report World Connections Committee (WCC) Report 06 Co-Dependents Anonymous Service Conference Countries Where CoDA Exists This report reflects the World Connections Committee (WCC) support of the growth and development

More information

GERMANY. WCPT COUNTRY PROFILE December 2018

GERMANY. WCPT COUNTRY PROFILE December 2018 GERMANY WCPT COUNTRY PROFILE December 2018 GERMANY NUMBERS 160000 140000 120000 100000 80000 60000 40000 20000 0 Physical therapists in the country Members in MO 21,502 136,000 Total PTs in country 800000

More information

LAB 4 Experimental Design

LAB 4 Experimental Design LAB 4 Experimental Design Generally speaking, the research design that is used and the properties of the variables combine to determine what statistical tests we use to analyze the data and draw our conclusions.

More information

Is there a relationship between Countries' Human Development Index (HDI) level and the incidence of tuberculosis?

Is there a relationship between Countries' Human Development Index (HDI) level and the incidence of tuberculosis? Is there a relationship between Countries' Human Development Index (HDI) level and the incidence of tuberculosis? Introduction Tuberculosis is a serious disease caused by the bacterium Mycobacterium tuberculosis.

More information

Overview of drug-induced deaths in Europe - What does the data tell us?

Overview of drug-induced deaths in Europe - What does the data tell us? Overview of drug-induced deaths in Europe - What does the data tell us? João Matias, Isabelle Giraudon, Julián Vicente EMCDDA expert group on the key-indicator Drug-related deaths and mortality among drug

More information

Chapter 11. Experimental Design: One-Way Independent Samples Design

Chapter 11. Experimental Design: One-Way Independent Samples Design 11-1 Chapter 11. Experimental Design: One-Way Independent Samples Design Advantages and Limitations Comparing Two Groups Comparing t Test to ANOVA Independent Samples t Test Independent Samples ANOVA Comparing

More information

PRODUCT INGREDIENT RELATED HEALTH CLAIMS ASSESSMENT for NU SKIN PHARMANEX PRODUCTS in EUROPE

PRODUCT INGREDIENT RELATED HEALTH CLAIMS ASSESSMENT for NU SKIN PHARMANEX PRODUCTS in EUROPE PRODUCT INGREDIENT RELATED HEALTH CLAIMS ASSESSMENT for NU SKIN PHARMANEX PRODUCTS in EUROPE This document is established in support of the European distributors communication regarding the right use of

More information

EPIDEMIOLOGY. Accurate, in-depth information for understanding and assessing targeted markets

EPIDEMIOLOGY. Accurate, in-depth information for understanding and assessing targeted markets EPIDEMIOLOGY Accurate, in-depth information for understanding and assessing targeted markets WHY ACCURATE MARKET FORECASTS Start with accurate understanding Epidemiology is the study of disease patterns

More information

What s s on the Menu in Europe? - overview and challenges in the first pan- European food consumption survey

What s s on the Menu in Europe? - overview and challenges in the first pan- European food consumption survey What s s on the Menu in Europe? - overview and challenges in the first pan- European food consumption survey Liisa Valsta Data Collection and Exposure Unit What s s on the menu in Europe? Background Attempts

More information

Nutrient profiles for foods bearing claims

Nutrient profiles for foods bearing claims Nutrient profiles for foods bearing claims Fields marked with * are mandatory. Background Regulation (EC) 1924/2006 (Nutrition and Health Claims NHC Regulation) establishes EU rules on nutrition and health

More information

MADE IN MOVEMBER GETTING IT GROWN FOR INQUIRIES PLEASE CONTACT FOR MORE INFORMATION ABOUT THE ORGANIZATION VISIT MADE IN MOVEMBER 2014 PRESS KIT 2

MADE IN MOVEMBER GETTING IT GROWN FOR INQUIRIES PLEASE CONTACT FOR MORE INFORMATION ABOUT THE ORGANIZATION VISIT MADE IN MOVEMBER 2014 PRESS KIT 2 USA GETTING IT GROWN THE MOVEMBER FOUNDATION IS THE LEADING GLOBAL ORGANIZATION COMMITTED TO CHANGING THE FACE OF MEN S HEALTH. WE ACHIEVE THIS BY CHALLENGING MEN TO GROW MOUSTACHES DURING MOVEMBER (THE

More information

The first and only fully-automated, multiplexed solution for Measles, Mumps, Rubella and Varicella-zoster virus antibody testing

The first and only fully-automated, multiplexed solution for Measles, Mumps, Rubella and Varicella-zoster virus antibody testing Bio-Rad Laboratories BioPlex 2200 System BioPlex 2200 MMRV IgG Kit The first and only fully-automated, multiplexed solution for Measles, Mumps, Rubella and Varicella-zoster virus antibody testing Bio-Rad

More information

Project Meeting Prague

Project Meeting Prague Project Meeting Prague IO1 Assessment 9.11.217 CHRISTINA PADBERG ON BEHALF OF FRANKFURT UAS Current Status Assessment matrix was fully evaluated Experts have been interviewed, Interviews were fully evaluated

More information

SS3 Series. Controlled Avalanche Power Diodes. Features: Mechanical Data: SMC/DO-214AB. Page 1 28/03/06 V1.0

SS3 Series. Controlled Avalanche Power Diodes. Features: Mechanical Data: SMC/DO-214AB. Page 1 28/03/06 V1.0 SMC/DO-214AB Features: For surface mounted application. Metal to silicon rectifier, majority carrier conduction. Low forward voltage drop. Easy pick and place. High surge current capability. Epitaxial

More information

ERNDIM QAP for qualitative urinary organic acid analysis. Annual Report 2003 (Sheffield)

ERNDIM QAP for qualitative urinary organic acid analysis. Annual Report 2003 (Sheffield) Ms M Downing Dr J R Bonham Professor R J Pollitt Telephone (+)44 114 271 7404 Fax (+)44 114 276 6205 Department of Paediatric Chemical Pathology and Neonatal Screening The Children s Hospital Sheffield

More information

Where we stand in EFORT

Where we stand in EFORT Where we stand in EFORT Engaging with the new EU regulatory landscape for medical devices. Challenges & opportunities Brussel, Belgium April 6, 2018 Per Kjaersgaard-Andersen Associate Professor Section

More information