Socioeconomic Determinants of Water System Management

Caleb White Author

04/04/2021 Added

8 Plays

Description

Why do some systems have violations and others do not? What socioeconomic factors contribute to violations?

Searchable Transcript

Search:

[00:00:00.840]Hi, my name is Caleb White,
[00:00:02.280]and today I'm going to be presenting on the socioeconomic determinants of water
[00:00:06.780]system management. I'm a professional engineer with eight years of experience.
[00:00:10.980]I currently work for JEO Consulting Group and I'm in the department of civil
[00:00:14.850]engineering under the advisement of Dr. Libby Jones. So let's get started.
[00:00:20.190]So here's my poster and the font is a little bit small, so, okay, here we go.
[00:00:24.930]Of all water systems in the United States,
[00:00:28.590]60.5% of these systems had a violation between 2014 and
[00:00:33.330]2017. I want to ask the question, why,
[00:00:37.080]Why do some have violations and others do not? Violations fall into four
[00:00:41.310]categories,
[00:00:42.360]and the two that I'm going to focus on are: public notification and other
[00:00:46.260]category, as well as the monitoring and reporting category.
[00:00:49.620]These I'm going to call management violations.
[00:00:52.110]And that is because these violations are responsibility of the operators and the
[00:00:56.730]staff who report to the state or could their consumers. And as you can see,
[00:01:01.050]a large percentage of these systems have these violations.
[00:01:04.800]It gives us a broader sample size to work with. Data sources.
[00:01:08.820]I'm using the EPA echo database pulling data from 2014 to
[00:01:13.200]2019. Socioeconomic data comes from the American Community Survey.
[00:01:17.190]five-year estimates. And from USDA,
[00:01:20.910]rural urban codes.
[00:01:22.920]I took this data and I matched it based on city identifiers only.
[00:01:27.720]Previous studies had used County level data to associate with a system.
[00:01:31.890]I wanted to look at just the city.
[00:01:33.540]Instead of looking at Lancaster County for Lincoln,
[00:01:36.660]I want to look at the Lincoln data itself.
[00:01:39.990]Some summary statistics: system size,
[00:01:42.870]I'm only looking at the small and very small systems.
[00:01:45.360]I want this to be applicable to the majority of the systems in the
[00:01:49.590]United States, which do fall into that smaller, and very small category.
[00:01:54.210]Ownership type: I'm only looking at local native American and public private.
[00:01:58.110]These are going to be the systems that where they operate is reflective of
[00:02:02.700]the community, the socioeconomic status of the community.
[00:02:05.490]So I've excluded state and the federal government ownership types.
[00:02:09.930]Water source. Most of these
[00:02:12.540]communities have a groundwater or surface water source. Violation type:
[00:02:17.010]Once again, most of these are going to the monitoring and reporting violations,
[00:02:21.000]followed by acute PNO or,
[00:02:25.680]public notification or other, and then non-acute type.
[00:02:30.090]Looking at statistical methods, I'm going to use binary
[00:02:32.550]logistic regression for these public notification and other,
[00:02:36.510]or PNO and monitoring and reporting violations using IBM SPSS
[00:02:41.250]27. First thing ran the analysis, eliminated some outliers,
[00:02:46.350]based on guidance there. Then also reviewed for multicollinearity.
[00:02:51.600]You will not see the socioeconomic for percentage of white [race]
[00:02:56.280]in the model. And that is because that was collinear for African-American or black.
[00:03:00.970]So I haven't selected that.
[00:03:02.160]African American included in the model.
[00:03:06.670]Results, 60.2% overall accuracy.
[00:03:10.800]The Hosmer-Lemeshow passed, the R-squared shows slight correlation,
[00:03:15.600]and then these variables in the equation I used to cut off limit of 0.2.
[00:03:20.250]Here are the results. I'm going to show these tables here for your information,
[00:03:23.970]just to look at it. And you can pause to take a look,
[00:03:26.760]but let's jump to the conclusions here.
[00:03:28.380]Socioeconomic factors are statistically significant,
[00:03:32.850]but I believe they are a fairly small contributor to the number of PNO/MR
[00:03:36.210]violations.
[00:03:37.620]Some of the effects that we saw proximity to a Metro area is associated with
[00:03:42.330]more of these violations, odds ratio of 1.2 to about 1.3.
[00:03:47.430]Rural areas are not statistically significant
[00:03:49.770]and therefore didn't show up in the model,
[00:03:51.600]This disagrees with the analysis of Virginia water
[00:03:56.100]systems. and so does the second one. So groundwater is associated with lower,
[00:04:01.080]PNO/MR violations, odds ratio of 0.663,
[00:04:05.610]which is a little bit interesting because,
[00:04:07.890]surface water tends to be more hands-on [operation].
[00:04:11.610]also some other conclusions, as far as racial populations go,
[00:04:15.210]the ones associated with decreases in violations were Asian,
[00:04:18.840]Hawaiian and Pacific Islander,
[00:04:20.700]and also black African is agrees with Schaider et al, 2019.
[00:04:25.740]Latino
[00:04:26.790]has an odds ratio showing an increase of 1.02 for Native American,
[00:04:31.320]also an increase. Once again,
[00:04:33.060]agreeing with some of the literature We also saw increases
[00:04:37.290]with percentage in poverty, which would make a little bit of sense here.
[00:04:41.550]but then also see a decrease, for systems with less population mobility.
[00:04:46.410]So if more of the system was in the same house as they were last year,
[00:04:50.790]then you're going to see lower the odds lower for violation,
[00:04:54.360]which that kind of makes sense, right? As people stay there longer,
[00:04:56.540]they're going to be more and more invested in the community.
[00:04:59.970]As far as public systems versus the public / private public systems by far had a
[00:05:04.050]much larger or higher odds ratio, almost 1.5,
[00:05:08.340]which was about the highest in the table there.
[00:05:10.980]So thanks again for stopping by.
[00:05:15.960]I'd love to answer any questions that you might have. Thanks.

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

Socioeconomic Determinants of Water System Management

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments