search

jcochanc / AHS

The American Housing Survey (AHS) has data on housing characteristic and demographic characteristics of householders. Our goal is to use the tools of PHP 2650 to develop a study question and examine the data.
0 stars 0 forks source link

Develop a Question #1

Open jcochanc opened 5 years ago

jcochanc commented 5 years ago

Our task is to examine the data available from 1997-2013 on https://www.census.gov/programs-surveys/ahs/data.html. Keep in mind that the questionnaire is updated. Any question we develop should have the variables required for answering in the span of years we are examining.

jcochanc commented 5 years ago

Hi Carolina and Linde, I had an idea: what about studying gentrification?

We discussed creating an index. It looks like this was done in LA - we could format the index and test in Providence so that it is relevant. Of course, we don't have to be limited to PVD - just an idea. For reference, this is the index developed for LA: https://www.arcgis.com/home/item.html?id=57e9231c3bd34d44ae49b309b0cb440e

lindechen commented 5 years ago

Professor Liu has OH on Thursdays 3-4:30pm. Before we commit to a research question and start discussing methodologies for analysis, I think it'll be a good idea to run our ideas by the Professor. Would you both be available to meet sometime next week, perhaps before Thursday?

lindechen commented 5 years ago

Building off of what @CarolinaVelasco proposed in her email, we could create multiple indices, each representing a different aspect of housing quality: environmental and health related factors, as was already mentioned, as well as neighborhood features, accessibility (commuting), etc.. Here is a codebook of the AHS that I found to be helpful: AHS mini codebook (see possible 'Topics' and 'Subtopics').

As for the research question, we may expect upward mobility to be an issue facing those without accessibility to good quality housing, potentially resulting in financial burdens, stagnant communities and weaker ability to set their children up for success. Variables indicative of this include: household income, delinquency/eviction, whether or not there is a person living with disability in the family, etc.. I'm not familiar with PCA or other dimension reduction techniques, but is it possible to create an index for the outcome as well? If so, then perhaps a financial or social/familial burden index or a mobility index (which would include education level).

Here is a paper on housing quality that combines measures of different dimensions of the quality of a household's housing situation into a summary index value: Causes and Indicators of Housing Quality.

lindechen commented 5 years ago

I like the gentrification idea as well, @jcochanc. Can we somehow tie this in with the housing quality topic via neighborhood characteristics?

CarolinaVelasco commented 5 years ago

My proposition is to build a Housing Quality Index. The two attached articles talk about the ZADEQ variable that was created by the AHS. ZADEQ captures housing adequacy but, adequacy is a narrower concept than quality. Housing quality is not only important for health reasons, but also for the housing market. The AHS is the most comprehensive national housing survey in the US, so building a quality index from this data has great implications. We can look at variables that may make the housing protective to environmental factors (air-conditioning, basement, flood plane, etc.) and that could impact health (sewage, water, presence of mold, pests, etc.).

As for the outcomes of interest, there are some possibilities. There is housing turnover in the AHS, which complicates the exploration of education attainment of children (i.e. we cannot assume a cohort study design). Instead, we could examine whether the children's grade is appropriate to their age. We could also look at the household head's education attainment and the race of current residents. That is, holding housing tenure constant, we can see whether specific scores of the housing quality index signal disadvantaged / minority populations.

I'd be happy to discuss this further in person after the break! It'd be also good to discuss our ideas with Prof. Liu.

Have a great rest of the break!

Best, Carolina

AHS - A measure of (Poor) Housing Quality.pdf AHS - Housing adequacy and quality as measured by the AHS.pdf

jcochanc commented 5 years ago

Please correct me if I am wrong; it sounds like we have a two-step process:

So I think we need to first scrape the data related to these two steps. Can we start a thread here that discusses the actual variable names and decide on what years we will analyze?

lindechen commented 5 years ago

The proposed question: Does housing quality help us predict the head of house's education level?

This question is relevant for policy makers, governmental departments, social groups, etc..