ICPSR in Education – Teaching, Learning, & Research Experience!

ICPSR in Education – Teaching, Learning, & Research Experience!

We’ll get started on the presentation. Again,
welcome to ICPSR in Education. Today we’re going to spend a little bit of
time talking about a number of different kinds of tools that ICPSR has
that can be used in teaching. Even though we titled the session “in
Education” the focus here really is on undergraduate education. So, we know
that some of these tools are used in graduate education as well as in K-12,
but the kinds of examples that I’ll be giving are from undergraduate education.
In large part because that’s the audience that I’m most familiar with,
so that’s what I can talk about sort of off the cuff. So I’ll talk a little
bit about resources that we have that were designed specifically for teaching. We have some “Help”… help in quotation marks there… for both
students and teachers. I’ll show you some of those. We have some… I think that
the tools that ICPSR has for finding and evaluating data can actually work really
well in the classroom. And so I’ll give you some examples of ways that you might
use some of those tools in teaching, as well. Then lastly, I’ll say a bit
about opportunities for students including the Summer Program in
Quantitative Methods of Analysis and the summer internship and the paper
competition. So get ready to learn about how to get your students using data and
hooked on research so that at the end they can take advantage of some of
these opportunities. Some of you have been in a couple of the
other webinars so far and so you know that we have redesigned our website. The
the tools I’ll be talking about first are available on the Teaching and
Learning with ICPSR tab. So I showed you, on this slide, how to get there from the homepage as
well as on any of the internal pages. So we are still the orange site, so orange will be the the
color of the day. It’s like Sesame Street, brought to you by the number three and
the color orange. The first resource that I want to talk about is our set
of Data-driven Learning Guides. These are materials that were
constructed after talking to teaching faculty and finding out that they really
wanted to use data in their classroom, but found that it was difficult for any
number of reasons. For example, we have an awful lot of data, so it’s hard to
to choose what works well in class or and sometimes the option is a workbook
that comes with a particular textbook and faculty members said that they
didn’t really want something like that because they felt like it would change the whole
course. They felt like they had to use the whole thing if it were there and
that they had students buy it. So what I’m going to show you is what we
developed in response to some of these concerns that faculty had. The
Data-driven Learning Guides are meant really for introductory level social
science courses, so it might be an intro public health or intro sociology. They
can be used in other courses, but the goal of the Data-driven Learning Guides
is to use data to explore some social science content rather than statistical
content. The Data-driven Learning Guides are in a standardized format so
they’re easy to drop into class sessions. And you know if you look at one and you
like it, then if you look at any of the other ones you’re going to get the same,
that same format. If you don’t like it, it also means you’re going to get that same
format and if you don’t like it I hope you’ll let us know why and we can fix. I made the decision not to jump out to
our website because we actually have quite a few tools that I’m going to
try to cover during this session, but because of that you don’t get to see
the list of the topics that are covered in the Data-driven Learning Guides, but I
took the titles and put them into the word cloud, so you can tell that we
cover a whole bunch of different kinds of topics. Anything
from voting behavior to attitudes about different topics to euthanasia and
gambling, attitudes about science, technology, engineering, and math
curriculum, all kinds of different things. We currently have just over 50 guides
and they’re, for the most part, centered around history, political science,
psychology, social psychology, and sociology, in part because the people who
have been working on developing them those are some of their background areas.
We’re working on getting some new content in this fall. We should have,
I think, four new Data-driven Learning Guides including one on astrology and
how we can use survey behavior data about… survey data about people’s
behaviors to look at how they might fit with their astrological sign.
And I won’t give away the story at the end about whether it fits or it
doesn’t. So I thought since we’re in the
election cycle here that I might talk a little bit about Data-driven Learning
Guides with “Attitude”. In other words, these are the ones that
are tagged with the term attitude. One of the topics that has been pretty hot in
this cycle has been immigrant policies. So I chose the one here on
American identity and immigrant resentment. Each of the guides, as I
said, are structured exactly the same way, so you’ll have these same sections that you see on the
left-hand side. The Goal & Concept section is just, the goal is a statement
or two that allow you as an instructor to look at the resource and say, “okay
yeah it’s worth my time to take a little bit deeper look” or “no this won’t work
for what I want to teach”. The Concept section, because it’s intended for an
undergraduate audience that’s not as familiar with social science, we try
to model good scientific practice throughout the guides. So the concept
section talks both about why social scientists might be interested in this
particular concept and how they would study it. So we end with examples of some
research questions. Again, the hope is that as students use these kind of
exercises in introductory level classes… I think of them as our gateway drug
to data, they start to think like researchers and by the time they get to
that senior capstone class and are given a dataset and told to do a
project with it they’ll be able to think like social scientists because
they’ve started to see the kind of example research questions that social
scientists would would investigate all the way from their introductory class on.
The Dataset section talks about why the particular data set that was chosen is
good to study this concept. The Variables section lists each of the
variables that were used in the exercise and if there were any manipulations done,
any recoding of categories, or any anything like that we note that in there
as well. Again, particularly in the era of open science, transparency, and
replicability. We try to to show students exactly what we’ve done. The Application
section is the heart of the guide and that’s what this chart on the
right-hand side is for. It comes from the Application section of the guide and within that section
there’s text and it asked questions of students and then
they can click links for pre-specified analyses. So this is a graph from one
of those pre-specified analyses in this immigrant guide. So it’s a table
looking at people who are resentful of immigrants on a one-to-three scale and
whether they think that the number of immigrants should be increased, decreased,
or left the same. So you can see that you get the cross-tabulation table and the
bar chart that accompanies it. The good thing is that faculty members
said that they wanted to make sure that all students were seeing the same thing.
They didn’t want sort of any room for error, for students to percentage the
opposite direction, for example. So these are all, as I said, pre-specified so
everyone is looking at the same thing. Because they’re pre-specified you can also
just take screenshots of the charts and drop them into your
lecture, you don’t actually even have to use the whole exercise if you don’t want
to. Lastly, the Interpretation & Summary
section and the Bibliography are just what they say. The Interpretation
section goes through the questions that were asked in the Application section
one more time, and then as students click through that, they can see what we think
the answers are. We start with some general things to think about for each
guide, like is there a lot of missing data? Is the sample that was
responded to the particular survey or whatever dataset is being used is
that representative? And then we also work through each particular statistical
test that’ss used in the guide to remind them how to understand each one. And I
should note that the guides are also categorized by level of statistical
sophistication. I can never say that. That is one way that you can filter out guides as you’re searching for ones to use in class, as well. So it’s by the highest
level of sophistication. This one, for example, the immigrant
resentment, has some multiple regressions in it and so it’s tagged as
multiple regression. If you’ve got one that’s tagged as cross tabulations, you won’t have anything more complex than that in the guide, but you can see
this one’s tagged with multiple regression and we do have a
crosstab. So again it’s by the highest level of sophistication so that you can
use them in a sequence if you want to teach students the different
statistical techniques. The Bibliography section includes resources
that are in ICPSR’s Bibliography that relate to the topic of the guide, so
it doesn’t just repeat the bibliography that is shown with the dataset that’s
used in the guide. So they have to be content-wise related to the guide, as well. The next tool that I would like to talk about is and what we call our Crosstab
Assignment Builder. If you know anything about ICPSR, you know that we live in a
world of acronyms, so at least this one’s pronounceable. So the Crosstab Assignment
Builder, or the CAB tool, is again found on that Teaching and Learning with ICPSR
page. It’s under the classroom exercises. This one was designed to be a
step from that gateway drug, Data-driven Learning Guides, between that and
independent analyses for students. The idea behind the Crosstab
Assignment Builder is that an instructor goes through and chooses a
dataset, and then chooses the specific variables that he or she wants
students to be able to look at, and then the students have the flexibility to
build the tables and interpret what they’re being shown. The key thing to remember about the CAB
is that it’s only available through the Teaching and Learning page. There’s a
little bit of confusion because it’s built on our online analysis package and
we have the simple crosstab tool with that. So in order to keep them separate, you have to get to the Crosstab
Assignment Builder through the Teaching and Learning part of the ICPSR
website. So once you’re there, you’ll be asked to click on a list of studies that
can be used with the Assignment Builder tool and then you’ll see on the
study home page, like I’ve got at the bottom of the screen, Build a Crosstab
Assignment. That’s how you know you’re in the right spot. I created an assignment here, the link works, so if you wanted to type that
in and follow it, you can see what a student would see. So I’ve set up the
table using the Current Population Survey, the Volunteer Supplement. And
I just created this example as a way of showing just a really limited
number of variables. I set it so that the percentaging is always on the
columns and so when I did that I went through and made sure that the variables
that I would want my students to think of as independent variables are
available for the columns and the dependent variables are available in the
rows. It may be a little bit difficult to see on the screen shot, but the tool also allows you to include recode syntax. So here the CPS has
quite a long race breakdown, I think there’s something like twenty four
categories. I wanted, for use in a cross tabulation table that’s a little bit
hard to interpret, so I wanted to collapse them down into three. So you can
see that the recode syntax is included there, as well. If you would go through and create this,
when you hit Save and Review at the top you’ll get a link the one that I
have on the slide. And it’ll take you to a screen that looks like this second
screenshot. This is what the students see. So they get… only on the left-hand
side you can see it says row column control. I didn’t select any to be
used as controls so none of them show up there. But you can see
that the list of variables for rows and columns, if I had taken picture when
that list had were expanded you would see that it’s only about five variables
for the rows and maybe 10 for the columns. So it’s a very small subset of
the Current Population Survey variables, but then students can use them and
look at, in this case, it’s whether the person has done any volunteering
by sex. So they can look at whether males or females are more likely to volunteer. We also have some exercise modules.
These are things that are… the two tools that I showed you so far are things that are
easy to drop in and use during one class period or even a portion of a class period.
The exercise modules are a little bit longer and they tend to be
most effective when all of the sections are used together. So these can be
used like as a term paper kind of project. Each of the modules was
designed by a faculty member and used with their own students, so we know that
they work. We also have some links on study home pages within the
publications list, to exercises that are housed elsewhere and I have some
examples of those in a minute. So the first exercise set that we have is called Exploring Data through Research
Literature and this one is based on the Bibliography. This was created by
Rachel Barlow who is a Research Librarian at Trinity College. Remember the
assignment that I mentioned earlier about, “here’s a dataset, go do
something with it” in a senior capstone class? She was tired of having students
come to her and say, “My teacher said I need to use the General Social Survey.
What can I do with it?” And so she built this series of exercises as a way of
getting students to answer their own questions. It’s a great exercise set
because it allows students to see that social research is not done in a vacuum. So the idea of the exercise is that
you, as a faculty member, choose a focal article from the Bibliography of Data
Related Literature and it takes a little bit of time on your part to make sure
that the article has enough ties to it that they can answer the rest of the
exercise, but the other kinds of questions are that they… that
students then take the focal article and they find one that maybe on the same
topic but by a different author, or on the same topic using a different
dataset, or maybe by the same author but on a different topic. So it also
shows students that the topic that they choose for that senior project does
not have to be the topic that they study for the rest of their career. The
next exercise module that we have is Investigating Community and Social
Capital. This one was developed by Lori Weber, who is a Political Science
professor at Cal State Chico and she uses this to structure her Research
Methods course. The activities take Robert Putnam’s “Bowling Alone” book and
work through it by teaching students to use a codebook for operationalization
and hypothesis creation. And then it helps them
think about specific research concepts and actually work through the analyses
that he includes in his book. And then she has some open-ended things to allow
them to take it a step further and explore on their own. Lastly, this one may be of some interest, again
because we’re in this presidential election cycle, but this one is Voting
Behavior the 2012 Election. We will have a 2016 version of this, but that won’t
be until at least the end of the summer because the data aren’t out yet. The data aren’t even fully collected,
yet. This Voting Behavior is based on the American National Election study
and it’s put together by two faculty members, Charles Prysby and Carmine Scavo. Prysby has been doing this since the mid-1970’s. This Voting Behavior module is actually part of what was called the
SETUPS: the Supplemental Empirical Teaching Units for Political Science. So
if you’re a political science faculty member and you remember using those back in
the day, this is the 2012 version of that. What I love about this
set of exercises is that the authors do a great job of teaching students both
about voting behavior and different patterns and about survey research, in
general. So they have additional readings and information about both of
the information tracks in the resources. So you can use it if
you’re not teaching political science but you are teaching survey methods, you
can use it for that, as well. I like that they kind of give equal billing to both.
I mentioned that we have links to other resources that are held elsewhere.
So if we know about resources that are built on data that we hold, we will
link out to them. So here is an example of a screenshot that includes
exercises using the General Social Survey which is one of the more popular… most popularly downloaded ICPSR studies. On the General Social
Survey related literature of bibliography, you’ll find
links to these variable or these exercises by Ed Nelson who is a retired
faculty member in Sociology. We also have some links on a few datasets that we hold that the Association of Religion Data Archives
also hold. They have fabulous teaching materials. Their scope is fairly
specific in that they deal only with data that has something to do with
religion. I should mention that shouldn’t be a reason for you not
to look at their things, because studies that have religion… that have good
measures of religion also have a lot of other things in it. So they have
questions about, “Where you get your information about science?” or “Belief in
extraterrestrials” is one that I’m always surprised to find on their
website. But they have some teaching tools that are built on their datasets and in any of the cases where we also hold the data, we link right back to
the ARDA. There’s also a series that was put together by a faculty
member in Political Science, Denise DeGarmo and she teaches International
Politics and spent a summer at ICPSR putting together a list of
different datasets, and topics, and just different resources that can be
used to teach International Politics. So this is by far not the only
international data that we have, but it’s… she went through and sort of curated
this collection as being data that she would use to teach particular topics in
an undergrad International Politics kind of class. In addition to the tools that
we’ve built specifically for teaching, some of our archives have tools, as well.
The Early Education Research Connections, Child Care and Early Education Research
Connections project has modules to teach pre-service teachers about using
data and the importance of understanding data so that they can understand sort of
what’s happening in their classroom and in standardized testing and things like
that. So this one is found on the Child Care and Early Education website
and I’ve tried to circle where these are. The National Archive of Criminal
Justice has a terrorism resource center and within that resource center are
some lesson plans and learning guides for teaching about terrorism. Not in the ISIS way, but about
understanding terrorism. They also have, and this is
an older screenshot because I didn’t have the link when I was putting the
presentation together to the new one, but coming soon they have a guide
that works with the National Crime Victimization Survey and teaches
students about the importance of replication and then works students
through to come up with the target numbers that the Bureau of Justice
Statistics publishes each year about violent crime and property crimes.
Lastly, one of our newest Archives, the National Archive of Data on Arts
and Culture also has a series of infographics and quick tables
that are quick, easy ways to bring data into the classroom. I like to start
class with a usually with a graphic depicting some form of data that’s
been in popular press. This would be a great source of that kind of
information that you could just use as a discussion starter
in any class. I mentioned that we have some “Helps” for students. We have a PDF on
how to read a journal article that we’ve been told is really helpful. And in
fact, some libraries have linked to it, have asked permission to link to
it. It sort of walks students through the different parts of a journal article and
what they should be looking for in each part. Those of us who read journal
articles on a regular basis know that you don’t need to read from the first
word of the abstract until the last word of the conclusions. But that if you’re
looking for particular things like “how do they measure these variables?” you look
in the Methods section. So that PDF helps students make sense of what
they’re seeing in a journal article. We also have a new resource on
interpreting SPSS output and I have a screenshot of that in a minute. We also, I would just mention here, that
we have a YouTube videos for common problems. So if a student is in a
statistics class and they’re having having trouble getting their software to
work, we do have YouTube videos on that on our YouTube channel, as well. This is a
screenshot of the… it’s actually a series of PowerPoint slides that work students
through reading SPSS output. For each of the statistical tests that are taught
in a basic introduction to statistics class. So it follows the format of a
textbook, in terms of which statistical tests are are shown and the order in which they’re shown. But it goes through and, as you can see, it corresponds an explanation with each
of the numbers that the students see in the print out. For things that
are more complex than frequencies they’ll also have the formula. So for
example, there’s a multiple regression slide that shows what the regression
equation would be for a particular regression model based on the output
chart. I mentioned that we also have “Helps” for faculty. We try to pay
attention to things that we find outside of ICPSR. This list is by far not
complete, but as we sort of come across things, we add them to the list because we know that sometimes that’s a good way to start looking for
other resources that might be useful for the classroom. So I mentioned the
Association of Religion Data Archives, they are my idols so I have to
point them out, they’re on the list. The Consortium for the Advancement of
Undergraduate Statistics Education, that cause link. They do a fantastic job
talking about best practices for teaching statistics and both for
teaching statistics overall and for teaching specific methods. They even have
a cartoon captioning contest, things like that. So there’s all kinds of
different resources there related to teaching statistics. The National
Numeracy Network also has a number of resources. It’s a good good place for
faculty members to go for professional development in terms of teaching using
data. A number of the tools that are on the
ICPSR website that are typically used for finding or evaluating datasets
can also be really helpful in the classroom. So we’ll talk a little
bit about the online analysis packets, bibliography, the search and compare
variables tool, study documentation itself, and our replication archive. The
online analysis package, SDA, is the tool that we use. That stands for Survey
Documentation and Analysis, and it’s out of the University of California at
Berkeley. We have just under 1,200 studies right now that have the online
analysis capabilities. You can find those by following the path on this
screenshot. If you go into the Find Data page under View All Studies, there’s a link to View All Studies for which online analysis is available. That’s how I know that there are, I believe… it’s like 1165… something like
that currently that have these tools. If you click on that, you’ll
see that there’s a wide range of topics. I went through and chose a New Family
Structure study as an example for this. That study had been in the news a
couple years ago and it’s just, I don’t know, it’s a fun one for me to look at, so
I use that as the example. One of the other kinds of datasets that are really
useful in teaching are the polling datasets. So ABC or CBS News Polls, New
York Times News Polls. Those tend to be smaller datasets, so that students
don’t get overwhelmed with the number of variables. They also tend to have
pretty timely topics. So they’re good sort of discussion starters for
class. I use those a lot in teaching, even teaching a statistics class,
because they’re nice sort of small data sets for students to get their head
around. Especially if you’re teaching about operationalization and/or levels of measurement they have a ton of Likert
scales often in them. So it’s a way to get students to see how to use ordinal level
data. If you choose a study like the New Family Structures Study, you can
see at the bottom there’s a… where it says analyze online, there are two options. The
simple crosstab frequency and SDA which is the full interface. I have a screenshot here of the
simple crosstab frequency tool, this one is a little bit more intuitive then the
than the overall SDA interface. The interface is great because students can
use it without having any SPSS, SAS, STATA knowledge, even Excel. They don’t
have to download the data. They don’t have to worry about variable labels and
value labels, those are already in the dataset. And they can do it from
anywhere. So this is particularly helpful for people who are teaching on campuses
with students who are working or it’s a commuter campus and so they’re not on
campus all the time to be able to work in a lab. They don’t need to be
anywhere specific, they can just… they can even do this from home. They can
use the online analysis tool. It’s also really good for in-class demonstrations. So you can use it if you’re talking about a particular relationship, you
can pull up a study and look at it right in class. I mentioned that the frequency
and crosstab tool is a little bit more intuitive, that’s what you’re looking at
in the screenshot. The rest of the package, while maybe not the prettiest
thing to look at, does have quite a bit of
functionality. So it allows you to do the kind of data management that you
could do in any other statistical package: recoding, creating new variables, that
kind of thing. It also allows you to do statistical
techniques up through multiple regression, logistic regression, probit
analysis. It’s also a way that we distribute some restricted-use data and
because of that some of the ICPSR archives have worked with the folks at
Berkeley to ensure that the results that you get, particularly with things like
complex samples, match what you would get in standard statistical software. So
if you’ve got students who are interested in doing research and your
University doesn’t subscribe to one of the main statistical packages and they
don’t want to learn R, I would definitely say on the intuitive level,
SDA is a little bit better than learning the syntax for R, so that’s
another option. I mentioned the Bibliography of Data Related Literature in the Exploring Research through Data… yeah Exploring Data
through Research Literature. Sorry had the right letters, wrong order. The Bibliography itself is great for teachers to send students,
particularly if students are looking for topics for research papers or
finding out what data have been used to study a particular topic. Anything that
they click on through the Related Publications links backward and forward
to the study homepage so they can see if they found an article that does
something that’s exactly what they’re interested in. They can look at more detail about the study itself, on the data on which that article
or book chapter or whatever was based. If it’s a journal article and it’s in a
database to which the library has full text access, you can also just get the full text
right through our Bbibliography. You don’t even have to go out to your school’s
library page. It does some some magic behind the
scenes and it knows what you should have access to. Yeah I think
that’s about all I wanted to say about the Bibliography, except that I really
encourage faculty members to have students start there, particularly if
they’re looking for data to do a research project. It’s probably the most
straightforward guide through our collection for students. We have a Search
and Compare Variables tool. It’s found on that Find Data page. One
of the options at the top is Search/ Compare Variables. This is something that is meant
to help you find data that have a specific set of variables within a
single dataset. I think it has great uses in the classroom, as well. So you
might use it in any class and I’ve got a screenshot of the tool on the next slide
so you can see what you would be looking at. But I would use
it in any class. We have data about all kinds of topics. So if you
find a survey topic, for example, even one of the news polls that looks at a
topic that’s related to what you’re talking about in class, have your class
answer the questions to that survey. This is especially easy if you
have clickers because then they can see the distribution and then show them the
distribution of the people who answered the survey. Around Ann Arbor, climate
change is sort of taken for granted. There are many conversations about
global warming, climate change, that kind of thing. Students are always amazed to hear that
not everyone thinks like they do. So in a class on public policy related to
climate change for example, it would be really useful to show students, “hey
look if you’re going out to work in this area,
there are some challenges that you’re going to need to overcome in that, not
everyone believes that climate change is even happening”. You can also use it in a class to show
different social trends, so for example I’m a Family Sociologist by
training, and so I use this tool to think about how we study family and look
at how the question of marital status has been asked over time. So now there’s
almost always a “living with partner” option that’s separate from “married”, but
that wasn’t always there. And so I have students think about, “where would those
people have put themselves before?” and “how does that change what we know about
those particular groups?”. For research methods that’s the example that I have on
this slide. I think this is a great tool to make topics that students find less
than enthralling in research methods sort of come alive. So you’re
teaching operationalization, which is how how do you measure a particular concept,
have them type in concepts and see what comes up. Or if you’re teaching about
sampling and you want to talk about the effects of having a nationally random
sample vs maybe a convenient sample. Find two questions that were similar
questions that were asked of those two different groups of people and look
at the response categories. Or if you’re talking about survey design, students
like to think about what happens if you give a choice in an attitude question of
“I don’t know” or “I’m neutral” how does that affect what you know
about response options? So the screenshot here is an
example of life satisfaction and it shows three very different ways of
asking the question and you get a little bit different view of society depending
on which one you’re looking at. So it’s a way to open up some of those
conversations with students. It’s also, if you have students create their own class survey and collect data
on campus, it’s a great way to start so that they
don’t have… they don’t have to come up with all their own questions. They can
just use questions that have been have been asked by others. Also in a
Research Methods class or in a class that’s related by content area, study
documentation that’s on our ICPSR homepage. For any
study, if we receive documentation that’s related to that study we will put that
out there. So maybe we’ll get interviewer instructions or like this first shot is
an executive summary of a research report. Those are out there along with the data.
So again this doesn’t require students to download the data. Reading the
executive report is a great way for them to get their feet wet in terms of
thinking about writing about numbers. It shows them that the kinds of things
that they might do in a job that’s related to research after college,
are things that they can do, they have the skills to do it. Often students don’t
want to admit that they think that they can do statistics because there
thinking about the higher-level kinds of things and that’s not always what it
requires. I mentioned that there’s methodological information. You
can use the documentation again, to look at sampling or time comparisons. You might also use it to talk about
different kinds of data collection efforts. So ICPSR primarily has survey
and administrative data, but the scope of our data collections is
increasing everyday. So we’ve got video data in the Measures of Effective
Teaching datasets or data collection. The Chicago Neighborhood Study, which the name was too long to fully
spell out, is another really interesting collection of data collections. So it’s
got administrative data, survey data, neighborhood data, all within the same
kind of collection. Those data are restricted, but the documentation is
not. So you can talk to students about what kind of data would be
necessary to answer a particular research question. We also have data management
information, this is a way to, if you’re talking about research ethics,
students often sort of snooze through that part. They assume that they know
everything or they’ve already seen the Zimbardo experiment video or whatever. Have them think about creating an informed consent document. We’ve got
information about that. It’s also a good place for graduate students who are
handling their own data to learn about how to keep that data safe, how to
ensure confidentiality, that kind of thing. I mentioned earlier the movement
toward open science and replicability of results. ICPSR has a number of datasets
that researchers have deposited because they’ve written a journal article or
some other publication and they need to deposit their data and/or their code.
This can be found again, under that View All Studies on the Find and
Analyze Data page. There’s View All Replication datasets. If you click that,
you’ll get a link to a list of datasets that we have not done anything with.
So having students work through those to find out whether they can actually
replicate the results is often an interesting experience. It’s not always
as easy to replicate as we would like to think it is. It also teaches them
about the decisions that are made behind the scenes. So “what did you do with this
missing data on this variable” or “how did you code that scale?”. In thinking about replication, there’s a
great partner in teaching good science, a project called Teaching
Integrity in Empirical Research out of Haverford College. They actually have a
protocol for teaching students how to create their own replicable data
processes, basically. Lastly, if you’ve used these tools and your students are
hooked on research, there are a number of opportunities for them to continue
learning. The ICPSR Summer Program in Quantitative Methods of Social Research
has courses on everything from Math for Social Sciences and Data Analysis
I and II up through some of the cutting-edge, latest and greatest
research methods. The dates for 2017 are already on their website, as are a
list of the typical courses that would be offered. Again, there are
undergraduates who take those courses, there are graduate students. I think
graduate students make up the bulk of the participants, but faculty members
also take some of those courses and people working outside of academia. We
have a summer internship program where students learn about processing
data, and they take courses in the Summer Program, and also create their own
research project. I have examples of those on the next slide. Then lastly,
if you have your students write papers using any of the data that we hold or
point to, please have them enter their paper in our student research
competition. There are cash prizes for these, and then the papers get published
on our website and in a special edition of the ICPSR bulletin. It’s a great thing for students
to be able to put on their resume. And that’s open to both bachelor’s and
master’s students. We have separate a competition for each. I said that I would show you what some of
our interns have done, and these files are huge so they’re apparently taking it
a little bit longer to load. We had a student using the Gates Millennium data
to look at the effects of the Scholar’s Program on whether students
aspire to continue in education after college. Another student looked at
the STRIDE data, looking at discrimination among sexual minorities. Two students used the High School and Beyond Survey. The first one looked at
drop out and sort of different predictors of economic independence
after students have left school for both males and females and sort
of what predicted that economic capacity the best. The other student looked at
the match between race and ethnicity of teachers and students to determine
whether that affected student achievement. So
that’s the end of what I had prepared. I haven’t seen any questions come in, but I
could have missed them. If you have questions, please do go ahead and send
them and I’ll try to make up some answers. I fear I may have put everyone to sleep. I’m not seeing any questions come in, so
I would just thank you for attending this session of Data Fair. This is
still sort of at the beginning. A couple questions have just come in.
Someone said that I answered all the questions that you had. I did try
to hone up on my mind reading skills at the beginning of the day. Someone asked,
“how do you get professors to buy into the tools that we talked about?”. One of
the easiest ways is to remind them that using these kind of tools into
their… integrating it into their teaching give students skills that they can
use on the job market. So more and more jobs are looking for people who can
think about and write about quantitative information. We get bombarded with that
kind of information every day and if students have those kind of skills
they’re much more marketable. And faculty are often… often resonate with
wanting their students to do well after college. Another way to get faculty buy-in is to
help them think about ways that these tools can parallel what they’re already
teaching. Maybe talk with them about a topic that they have trouble keeping
students’ attention, and each class has those kinds of topics, and think about
would an active learning exercise using one of these tools make that
content something that students are more interested in? We know that active learning and the use of data in particular, allow students to remember
content, substantive content, better. So sometimes that’s a way to get
professors to buy into them, as well. With that, I think I will end this
webinar. Thank you all for attending and I hope to see you at another webinar of
this 2016 Data Fair.

You May Also Like

About the Author: Oren Garnes

Leave a Reply

Your email address will not be published. Required fields are marked *