Assignment 2 – Pure replication and completion plan
计划paper代写 Reproduce and discuss descriptive statistics and main results of your OP Main result = main table or figure (Yes, you have to use your own…
- Reproduce and discuss descriptive statistics and main results of your OP
- Main result = main table or figure (Yes, you have to use your own judgment as to what the “main” table or figure is!
- Discuss issues you have run into thus far
- Outline steps for completion, including your planned extension
- Word count: 1,200 words
- Weight: 25%
- Total marks:100
Learning by Replicating
1. Introduction 计划paper代写
Briefly state (in 2-3 paragraphs) why the paper is interesting (both generally and as it relates to the economics literature), the key research question(s), the empirical approach, and the main findings. You can repeat some of the things you included in your presentation, but hopefully by now you have also built on that knowledge and gained a deeper understanding of the literature and paper itself.
Outline what you will discuss in the subsequent sections of your paper.
Describe the replication data and code provided.
For example, in a seminal study, Berger, Ho, Maitra, Phan, and Xu (2022, henceforth BHMPX) discuss a model for experiential learning activities in the postgraduate classroom.
BHMPX make their data and code available on the American Economic Association website (you can describe any complications in an appendix). The replication package includes X datasets, in a combination of comma-separated files (.csv) and dusty ancient scrolls. In the 5 household survey datasets, observations are at the household level (i.e., a household identifier uniquely identifies each household). The data file describing labor market outcomes is at the individual level and can be merged on to the household data. There is also a village-level data file, which has information on prices of common purchases. The authors include .do files to clean, label, and merge the data. Separate scripts run the analysis, and the readme file describes the process in detail. The dusty ancient scrolls are seemingly included by mistake as they are written in Aramaic.
3. Pure replication 计划paper代写
3.1 Reproduction & discussion
Reproduce the summary statistics from OP. Discuss any discrepancies (examine the number of observations, for example, to make sure that you are working with the same sample).
Identify a table or figure that constitutes (in your opinion) the most important or most interesting result in the paper. Note that there may be some overlap here with your presentation materials; that is fine, so long as you are rephrasing your statements in written language (a.o.t. presentation bullet points or spoken language).
Reproduce this main table / figure and discuss. You should aim to discuss both (i) the main findings and their interpretation as it relates to the hypothesis of the OP, and (ii) anything you learned / noticed / struggled with in replicating it.
⚠️ Do NOT include screenshots or copies of OP’s tables or figures in your paper. You should report the version that you have reproduced, with reference to the relevant paper from OP and any relevant table or figure notes.
⚠️ Do not worry too much if your numbers do not match exactly with OP. Comment on any major discrepancies and what you think the reasons are.
⚠️ Do NOT take screenshots or regression output from Stata. Your figures / tables should be “real” tables and high-quality (high-resolution) graphics.
⚠️ That said, do not focus too much on the formatting of your table at this point. It takes time to learn to generate beautiful tables (and figures).
3.2 Code and challenges 计划paper代写
Describe any issues that you faced with the code that the authors provided. Describe any challenges you have faced so far, and how you approached solving them. Reflect on what you’ve learned from the exercise thus far.
If you are unable to reproduce one or more of the OP tables / figures…
…describe the reasons why you are submitting your assignment without having reproduced the main results. Is it due to software access? Coding issues (for example, the code is too complicated or there is a substantive coding error that you have been unable to solve)? Mismatch of software version? Whatever your challenges, discuss what steps you have taken to overcome them and what the outcome has been. Give yourself a pat on the shoulder for persisting with this difficult task and reflect on what remains to be done.
If you have successfully reproduced relevant tables / figures…
…comment on your experience in general (after you give yourself a little high five!). Has it been a smooth process, or did you face issues along the way? If so, how did you go about addressing them?
4. Roadmap 计划paper代写
Review our lecture on the typology of replications and explain your planned approach. Using the terminology of Clemens (2017), how do you plan to extend the OP’s work?
Why is your proposed extension meaningful or interesting? What value does the proposed extension or re-analysis add to OP or to this literature? How are you advancing knowledge? Ideally, cite a high-quality reference from the literature in support of your plan.
If you plan to do a re-analysis, how do you propose to alter the OP’s specifications? You should write the OP’s original estimating equation as well as your planned modification and explain what you are doing differently. For example:
“In estimating the effect of OP’s intervention on poverty, I plan to re-estimate per-capita consumption using OECD equivalence scales. This may affect the estimates and interpretation of the study, since the control households had fewer children at baseline (as can be seen in the descriptive statistics in Table XY).”
If you plan to do an extension, what is your proposed sample/population? If you propose to bring in new data, you should provide details: source, survey year, cross sectional/longitudinal, unit of analysis (household, individual, firm, country, village). For example:
“OP’s analysis focuses on developed countries and I propose an extension that analyses developing countries. I will use comparable data for Bangladesh, Indonesia and Columbia, sourced from Gallup World Poll surveys, 2015-2017 (cite data source).” 计划paper代写
“I plan to extend OP’s analysis, which covers the years 2015-2017, by analysing more recent data sourced from Gallup World Poll surveys, 2018-21.”
Discuss any logistical issues you anticipate you might face, such as issues in accessing data, software etc.? In general, briefly discuss any problem you might be anticipating in implementing your plan and mention a Plan B if your Plan A fails.
- Complete pure replication & identify key challenges.
- Get access to the data I need for extension, sort out any issues, and get ready for data analysis.
- Familiarize myself with the data beyond the pure replication. 计划paper代写
- Continue refining methods needed for extension. What do you want to test? How will you test this hypothesis? Which statistical tests will answer your research question?
- Begin estimating results needed for extension.
- Interpret results from extension. Discuss what we know now that we did not know before my study.
- Begin writing up and discussing results.
- Reflect on my personal growth and learnings during the capstone.
Weeks 12- 13
- Continue writing.
- Complete first draft and move towards completion.