Instructions

The deadline for this Coursework is Wednesday 9th November 2023, 10am.

Upload your answers in a single file via Turnitin using the link on KEATS.

Include all the Stata commands you used to answer the questions either at the end of the file or in the text of your answers.

Answers should be accompanied by concise explanations.   Keep your answers short, around 200 words, and no more than approx.  300 words for any question (excluding Stata commands and tables). Overly long answers will be penalized (for example, more than 500 words per answer). Ablank Yes/No answer will not receive a full mark. Please explain your steps and answers as well and precisely as possible.

Presentation matters and good writing will be rewarded. We will award marks for neatly presented answers with well-formatted tables, equations, and paragraphs.

Working together is acceptable, however, the final write-up must be individual. Turnitin is run for academic integrity. Students are assigned different sub-samples therefore your estimates and interpretations could be different.

The use of Generative AI in the coursework

Following the rise of Generative AIs, King’s College London has created the King’s guid- ance on generative AI. Please read the guidance very carefully. Students are also re- quested to make a declaration in the Cover Sheet attached to the coursework on whether and how they have used AI for the coursework.

For our course, we believe that when appropriately used, Generative AIs can be valu- able to your learning process. However, it is paramount to recognize that Generative AIs are not databases of knowledge, and information generated from these tools may or may not be correct.  It is therefore imperative for students to be able to use their economet- rics knowledge acquired from our course to recognise and make use of Generative AI’s responses. Question 6 in this coursework will be about this matter.

The data

In this coursework you investigate the relationship between a student’s grade in econo- metrics and their lecture attendance, highschool achievement, part-time job, and parental education. The file cwdata2023. dta contains survey data from a UK university. The vari- ables are:

Questions: Answers must be accompanied by concise explanations

1. (30 marks) Construct a scatter plot of the exam grade and the average lectures missed per week. Is there a relationship between the two variables? Run a regres- sion of grade on skipped.  Interpret the coefficients.  Discuss briefly the statistical significance and whether the slope is economically meaningful.

2. (10 marks) Can we interpret the slope coefficient obtained in (1) as the causal effect of missing lectures on the overall grade? If yes, explain. If not, give an example of potential biases and explain the direction of the bias.

3. (15 marks) Do students with a part-time job (parttime = 1) attain a higher overall grade on average than those without a job?  Use appropriate Stata commands and statistical tools to answer this question. How about students with at least a parent with a university degree (heparent = 1)?

4. (20 marks) Add parttime, heparent, and averaghs as independent variables into the regression in (1). Describe what happens to the slope coefficient of skipped com- pared to what you find in (1).  Using your answers to (2) and (3), explain whether and why the coefficient of skipped is different from that in (1).  Interpret also the coefficients on the other independent variables.

5. (10 marks) Make a table of results for the regression estimates from Questions (1) and (4). Explain briefly which would be your preferred specification.  [Hint: Make use of the commands estimates  store and est  table, as seen in the tutorials.]

6. A question using information generated from a Generative AI

i. (10 marks) You can find on the next page an econometrics question and the answer generated by ChatGPT 3.5. Concisely and critically evaluate this answer using the econometrics knowledge you have learned so far.  [Hints: You can evaluate the answer point by point, paying attention only to its econometric content,not the quality of writing or cohesion.]

ii. (5 marks) Provide a short reflective paragraph (up to 100 words) about what you learned from this exercise. [Hints: There is no right or wrong approach to answering (ii.)]