Team Blue
What is a "Field Experiment"?
An experiment which is conducted in the real world instead of a laboratory. A researcher randomly assigns participants to a treatment and a control group. The researcher decides what the treatment is. In order to reach causation, a field experiment must assign participants randomly to the treatment and control group.
How do you interpret log coefficients?
percentage change, e.g. "a 10 percentage-point increase in the share of peak hours leads to a decrease in performance of 3.4 percent"
The abbreviation "ATE" in the De Grip & Sauermann (2011) paper stands for ... ?
Average Treatment Effect. The average treatment effect (ATE) is a measure used to estimate the causal effect of a treatment or intervention on an outcome. It represents the average difference in the outcome variable between a group of subjects who received the treatment and a group of subjects who did not receive the treatment.
What do we understand under "Informal Learning" in the workplace?
"informal learning in the workplace could be defined as the acquisition of skills through learning by doing as well as by watching other workers, taking instructions, receiving supervision or feedback from supervisors or co-workers and self-study" (De Grip 2015, p.2)
What is "exogenous variation"?
Exogenous variation in treatments describes that the assignment to the treatment is not affected by any factor inside the analyzed system, that is, no important variable, from the perspective of the research question, influences the assignment. In this situation identification of treatment effects is possible because assignment is not related with the outcome variable. The study by De Grip and Sauermann (2011) uses exogenous variations in training participation, by assigning individuals randomly to treatment/control groups to identify the causal effect of training on individual productivity.
What do we understand under "Formal Learning"?
Learning that is structured and takes place within a planned setting, like a classroom or online. It has clear objectives and goals that are defined by an instructor or trainer.
What is a "difference in difference" analysis?
This analysis method looks at differences in the differences of the outcome you are interested in. Sometimes, when a group is not exposed to the intervention, their outcome might change as well. In order to know the true impact of the intervention, we need to look at the changes in the outcome for both groups (check if groups are similar!). DiD is therefore an analysis method that compares the performances within treated and controlled individuals AND across time. → DiD controls for unobserved heterogeneity that is fixed per subject (Handbook)
What is a Regression Discontinuity Design?
The RRD aims at creating a treatment group and a control group that are almost exactly the same, but due to a very small difference belong in different groups. There is a cut-off point and effects are measured before and after the cut-off point, in this case the treatment (training). The "jump" from pre-cut-off to post-cut-off is considered the causal effect of the treatment.
What are "fixed effects"?
Fixed effects is an analysis method that can deal with the problem that individuals can differ from one another based on factors we cannot observe. This is called heterogeneity. By using the fixed effects analysis we can capture and account for the unobserved factors. Doing this helps you getting a better understanding of how individual characteristics influence the dependent variable we are interested in. (Handbook). You need longitudinal data, include all the relevant observable factors and unobservable factors need to be fixed and not time variant.
No causality, unobserved heterogeneity is still possible.
In De Grip and Sauermann (2011) we find individual fixed effects and period fixed effects.