Question

·      Define and apply concepts of temporality, association, and direction

·       c. Explain the role of counterfactual reasoning in causal inference

·       d. Describe how randomization usually enables causal inference

·       e. Understand (in general terms) threats to causal inference even under randomization: failures of

·       excludability and non-interference.

Answer 1

A.Time and temporality has, for the most part, evaded thorough examination and is often sidestepped or assumed to be a non-contentious issue in frameworks that seek to explain organizational change. Temporality (past, present and future) contrasts with atemporal and tenseless conceptions of time where change is viewed as a series of ‘now’ moments in which the past and future are represented as social constructions that serve to make sense of an ongoing present. In the field of organizational change, time remains integral but opaque in theorization and implicit in the explanations captured in macro planned and episodic models characterized by linear temporality in which changes progress through a series of sequential stages, through to the more micro explanations of emergence that focus on continuous reconstituted becoming in changing organizations. This poor conceptualization of time requires attention to further develop theorization and enable researchers to engage in richer empirical work. The article unpacks conceptions of time that underpin change theories and suggests that the concepts of temporal orientation, awareness and accommodation can be used to open up and reflect upon temporality in generating a wider debate and furthering discussions on the place of time in understanding processes of change in organizations.

B. Counterfactual reasoning plays an important role in causal inference, diagnosis, prediction, planning and decision making, as well as emotions like regret and relief, moral and legal judgments, and more. Consequently, it has been a focal point of attention for decades in a variety of disciplines including philosophy, psychology, artificial intelligence, and linguistics. The fundamental problem facing all attempts to model people’s intuitive judgments about what would or might have been if some counterfactual premise A had been true, is to understand people’s implicit assumptions as to which actual facts to “hold on to” in exploring the range of ways in which A might manifest itself.Many formal theories of counterfactual reasoning are inspired by the model-theoretic accounts of Stalnaker (1968) and Lewis (1973). Minor differences aside, both crucially rely on a notion of comparative similarity between possible worlds relative to the “actual” world of evaluation. Simplifying somewhat, a counterfactual ‘If had been A, would have been C’ (A C) is true if and only if C is true at all Aworlds that are maximally similar to the actual one. Stalnaker and Lewis account for various logical properties of counterfactuals by imposing conditions on the underlying similarity relation, but neither attempts a detailed analysis of this notion. Much of the subsequent work on modeling counterfactual reasoning can be viewed as attempts to make the notion of similarity more precise

D.Randomized experiments, like any other real world study, generate data with missing values of the counterfactual outcomes.However, randomization ensures that those missing values occurred by chance. As a result, effect measures can be computed –or, more rigorously, consistently estimated–in randomized experiments despite the missing data. Let us be more precise.Randomization is so highly valued because itis expected to produce exchangeability. When the treated and the untreated are exchangeable, we sometimes say that treatment is exogenous, and thus exogeneity is commonly used as a synonym for exchangeability.

E.This paper reviews the role of statistics in causal inference. Special attention is given to the need for randomization to justify causal inferences from conventional statistics, and the need for random sampling to justify descriptive inferences. In most epidemiologic studies, randomization and random sampling play little or no role in the assembly of study cohorts. I therefore conclude that probabilistic interpretations of conventional statistics are rarely justified, and that such interpretations may encourage ministerpretation of nonrandomized studies. Possible remedies for this problem include deemphasizing inferential statistics in favor of data descriptors, and adopting statistical techniques based on more realistic probability models than those in common use.

F.

Let’s consider this example. Suppose, you buy a car and the same car is not available to anybody else, because of lack of supply. That means, there is rivalry in consumption.

Additionally, some of your neighbours might not be able to afford a car altogether, because of their buying power. Hence, there is an excludability in the consumption of the service.

Now, if we replace this example with Street lights.

An additional pedestrian on a well-lit street does not add to the cost of providing the street lighting. Hence, there is non-rivalry in consumption of service.

Additionally, it is impractical to set toll barriers around well lit streets and charge for people to walk down them. Hence there is non-excludability in consumption of the service.

Due to non-rivalry and non-excludability, there are certain services where the market mechanism fails, either completely or in part. Hence, the government provides them.