1845	Assessment Initial Draft	Final Draft comments
Abstract	No abstract present.	Overall, a good abstract.  It could be strengthened a bit by spending slightly less time on background and more on the research question and design.
Research Question	The research question seems fairly clear and reasonable in the introductory section, with a focus on poor whites.  However, some parts of the draft seem to imply a different research interest, with a focus on blacks, seemingly reflecting a change in focus.  This should be made consistent.	The research question is well defined in the Introduction and sustained throughout.
Literature Review	Overall, the review of the selected literature is pretty good.   It could still benefit from organizing more in terms of the ideas rather than the authors.  And, given the topic, the literature review should have a bit more depth and breadth, which will come easier if it is organized more around competing ideas and findings. The literature review is mostly focused on research surrounding the black-white wealth gap, yet it is the gap between poor whites and other groups that is highlighted in the introduction and research question.  This should be consistent. Additionally, the proposal might consider some research on the consequences of wealth inequality, such as how wealth differences may impact individuals’ abilities to pay for college, retire, etc. Right now, the proposal assumes that the wealth gap is sociologically important (which is true) but does not talk much about why it’s important.  Of course, it is obvious why being rich is better than being poor, but perhaps not so obvious why the more limited differences between groups lower in affluence matter over time. The research might benefit by avoiding, or at least facing head on, a conceptual issue that appears in much research in this area, that is neglecting fundamental differences between wealth invested in homes (and vehicles in a bit different way) and "financial" wealth.  Wealth invested in a home influences one's standard of living, and we derive benefit or "value" from it independently of changes in its market price.  The market value of a home commonly becomes translated permanently into financial instruments only when the owners no longer need or want to own a home, presumably most often due to death or the deteriorating years leading to death, secondarily due to hardship or other circumstances inducing a move toward rental property.   The proposal need not go much beyond recognition of the issue at the proposal stage, but it could be a good idea to prepare for taking the issue into account as the research evolves. As the phenomena being studied varies significantly over time, the proposal should be careful about which research findings match which periods.  It does not work, for example, if we jump from citing a finding in 2010 to one in 1970 as if time were irrelevant.  Similarly, if assessing a contemporary conditions, it is problematic to rely on findings about how things worked a few decades prior; in such a case, find more recent research or state why the old findings remain relevant.  Some details: In varied sentences the text now makes authors the subjects; try to make the findings or arguments the subjects and mention the authors in citations or secondary clauses (e.g. "X often leads to Y, as shown by Z").  If the only, or primary, role of education is to influence earning capacity, then it should be discussed in the context of income, not later.  The presentation flow should complement the logical relations between the things being presented.	Overall, the literature review is broad and analytical, but it could be more effective.  Much of it is about the background story of wealth inequality, which is interesting but not that important for proposal (providing such a background can be valuable for a paper reporting the research, but in a proposal it is usually best to focus mainly on the literature that informs the research question and design).  We could use more review of material directly relevant to the research question.  the review still jumps somewhat awkwardly (and confusingly) between the black-white wealth gap and the low-income vs. high-income wealth gap.  The critical section on "What Influences Wealth?" would be much more effective if it had a reasoned, understandable organization.  The confusion probably relates in part to using "wealth" very abstractly.  We cannot explain wealth as an abstraction, only in some more concrete form.  For example, we can ask how wealthy people differ from those without wealth, we can ask how different kinds of wealth grow differently over time, we can ask how the social distribution of wealth changes over time.  Different questions about wealth make different causal processes relevant. Going forward, it might help to think through the ways that divergent influences on wealth have an impact over time and use this to place them into conceptual categories.  Perhaps start by clarifying that you are focusing on the family or household as the wealth holder rather than individuals (at least, this appears to be the intent).   Then consider differences between characteristics that refer to the capacity to create or accumulate wealth, the forms wealth can take, the ways that wealth comes into people's hands or that it grows, and the like.  While you are still gripping onto the goal of a descriptive research project, this is again a place where we see that description is problematic in the absence of causal reasoning.
Data & Analysis	The PSID offer good, but rather complex data.  Although the problems being raised can be studied effectively with longitudinal data, it is not clear that the research design involves using the strengths of panel data.  If the stress is to be on the comparison of group characteristics, comparing cross-sectional results across time could be an alternative strategy.  The proposal might benefit from consulting the paper Measuring Wealth and Wealth Inequality: Comparing Two U.S. Surveys (https://psidonline.isr.umich.edu/publications/Papers/tsp/2014-03_%20Measuring_Wealth_and_Wealth_Inequality.pdf). Since the proposal hypothesizes that geography plays a significant role in wealth redistribution, the research design might make use of a multilevel model with households clustered within states (or counties in the restricted dataset) to account better for clustered error terms.  This is a statistical issue beyond our immediate concerns, so this is just a note for to think about in the future.	The data sound fine.  The proposal would benefit by providing a clear description of the virtues and shortcomings of the data for this project.  Even if someone reading the proposal is familiar with the data, they will want to see that you are familiar with it.  Getting access to the restricted data for the geographical identifiers will be a plus. The proposal states that it will compare the circumstances of only low-income whites in 2007 and 2016 ("Approaches and Methods" section).  Unless I am missing something, it will make more sense to compare across periods for the entire population, so that you can compare the changes and apparent causes for low-income whites with those for other groups.  After all, the slowness of the wealth recovery for low-income whites is a comparative characterization.  This might involve doing separate, parallel analyses for the race-income groups you define, or it might involve an integrated analysis of all the data using income and race as variables.
Causal Interpretations	While the proposal characterizes the study as descriptive, it appears rather causal.  Why does the proposal appear to avoid causality?  Why does the number of interactions (p. 7)  prevent it from making causal claims? By itself, varying access to real estate by its value appears insufficient to explain variations in wealth accumulation.  Rather we would expect it must be varying access to real estate by value appreciation potential (note that the price of initially lower valued real estate may rise faster) or the relative appreciation potential of alternative investments (consider renters as a basis for examining this).	As further evidence against the characterization of the project as descriptive, the section Theory and Hypotheses is about explanation and causality (as it should be).  The causal reasoning spread through the proposal is okay for this stage in the project, but analytic refinement will be crucial for moving forward.  Essentially, the research involves several levels of causal reasoning, which is a bit complex, and will benefit from clear recognition of what needs to be done.  Look at this list of causal reasoning levels: (1) what causes families to accumulate more or less wealth over time?; (2) what causes the significant and persistent difference in wealth accumulation by race?; (3) what induced the long-term historical shift in the inequality of wealth accumulation?; (4) how did the Great Recession disrupt wealth accumulation?; and (5) why have low-income whites seemed to suffer the least recovery of wealth since the Great Recession?  While one can divide the analytical pie differently, this provides a rough sense of the big causal processes that lead to the final one that is the direct focus of the project.  It may be possible to carry out the project by considering only the direct questions (5), but realistically the quality of the answers to this question will depend to some degree on the answers to the preceding ones.
Research Contribution	While the research value is fairly self-evident, the discussion of its value could be a bit clearer.  Try to return to the earlier mention of limitations to past research, and suggest how this research will respond to that.  The direct advocacy for the proposed research does not need to list the expected findings (although it can if it helps make the argument).	The discussion of the project's expected value is fairly good.  Given your interest in connections to political issues, it seems odd that no connection is made to the perceived role of low-income whites in the last national election cycle.
Citations & Bibliography	Okay, but note that when using a full bibliography with footnote citations, even the first footnote citation of a publication often uses simplified (rather than complete) publication data.	The reference structure is okay, but has some sloppy flaws.  For example, look at the notes at the bottom of p. 8.  The citation for Pfeffer, Danziger, Schoeni has no bibliographic entry (presumably Fabian T. Pfeffer, Sheldon Danziger, Robert F. Schoeni.  "Wealth Disparities Before and After the Great Recession."  The ANNALS of the American Academy of Political and Social Science, Vol 650, Issue 1, pp. 98 - 123) and the PEW research report has the first author's (Rakesh Kochhar) name listed incorrectly by his first name in the bibliography.  Citations and bibliographies should be complete and without errors.
Quality of Writing & Organization	The proposal could use editing for minor grammatical and syntactic errors.  Also, some minor logical flaws are jarring (e.g. on p. 2, the initial phrases of the last 2 sentences of the first paragraph are inconsistent; on p. 3, 4th paragraph, last sentence, the final clause does not fit the earlier ones).	Again, edit carefully to avoid logical flaws.  For example, consider this sentence from p. 5: "Today the wealth gap for median white households has grown to be twenty times greater than the median income for black households."  Or on p. 10: "This prospect is especially significant given policymakers’ admitted lack of understanding of what interventions they used prevent the severity of the 2008-2010 foreclosure crisis."  Bringing readers to a halt as they try to decipher sentences leads to irritated rejections.
Priorities for Revising / Responsiveness to Feedback	This seems a solid first draft.  All parts could use some significant investment of time to improve the consistency, internal logic, and clarity of presentation, but the foundation is there to build on.	This version does a good job overall responding to earlier feedback.
Miscellaneous Notes	Sometimes when you say "determine", consider using "influence" or the like ("determine" suggests a cause is sufficient and possibly necessary, which you usually do not intend).
Proposal		A good research project, with the proposal having some minor flaws.
Class Overall		Excellent overall work in the class.