Assignment review

Modules/Module5/Mod5Home.html

Module 5 – Home

Comparing Models to Assess the Research Problem

Modular Learning Outcomes

Upon successful completion of this module, the student will be able to satisfy the following outcomes:

Case
Assemble the methodology chapter.

SLP
Use data and visualization tools to complete the background.

Discussion
Reflect on the dissertation task ahead.

Module Overview

Data analysis and presentation

The final stage of describing your research methodology is the development of a data analysis plan and outline for development and presentation of your findings, both to your research colleagues and to your research site. The analysis is neither automatic nor trivial. Data tend to be complex and interpretable in numerous ways, and there are specific rules about how different kinds of data can be handled and the kinds of inferences that can be drawn from them. Different audiences are interested in different parts of your study, and selecting the appropriate presentation formats for different audiences can have a critical bearing on the success or failure of your overall research effort.

Analyzing the data and organizing them for presentation is essentially the construction of an organized story told by the data. Thus, developing this part of your project requires that you understand what your story is and that you be able to tell it clearly and effectively. It’s critical to remember that your data, whatever their nature and however you collected them, are essentially living things, since they are abstracted properties of living organizations. As living things, your data deserve as much respect and attention as you would pay to any other living participant in your research. Understand that your data really want to tell you a story; all you have to do is listen clearly to them without preconceptions and without a prior agenda and the story will come through.

Qualitative data in particular, including most of what is usually derived from interviews, are by necessity going to be at least partially analyzed as you proceed. That is, as you reflect on each interview and what you learned from it, you will learn additional things to investigate and form certain preliminary conclusions that you may test in subsequent data-gathering. It’s important not to get yourself committed to any one set of results to early on, since later information may change or modify your earlier conclusions. But it would be naïve to pretend that you are not forming certain conclusions as you proceed. Quantitative data, particularly those derived from questionnaires and/or secondary analysis, tend to be analyzed after the fact. Typically, you only get one shot at collecting these kinds of data, so you need to be sure that they are really what you intend to collect.

Presenting your findings will involve at least two stages. First, you will be writing up an academic report that will constitute the last two chapters of your doctoral project. These chapters on findings and discussion will be appended to your previous proposal as your final project. This is pretty much an academic exercise, the kind you have had a lot of practice with to date. The second stage involves presenting your findings to your client organization, usually through or at least in conjunction with your project liaison person. This is likely to be a different kind of presentation, emphasizing more the practical results of your study and the recommendations, if any, that you have formed for the organization as a result of the study. This presentation is likely to be interactive, so you need to be prepared to answer questions about what you have done and what you think you found. You want to avoid any more academic language than you have to use and make the presentation as accessible to the people in the organization as possible. This part of your project is as important as the academic write up, and you need to think clearly about how you want to organize it.

Obviously, you cannot analyze data you do not have or present findings that you have not formulated yet. So, this part of your project proposal will be more hypothetical than other parts of your plan. Nonetheless, it’s important to think about these issues in advance and have some ideas about how you will approach the data and who will be interested in various parts of your results.

Privacy Policy | Contact

Modules/Module5/Mod5Background.html

Module 5 – Background

Comparing Models to Assess the Research Problem

Required Reading

EvaSys. (2018). How to effectively carry out a qualitative data analysis. Retrieved from https://www.evasys.co.uk/wp-content/uploads/2019/09/How_to_Effectivele_Carry_Out_a_Qualitative_Analysis

Reporting and discussing your findings. (2018). Retrieved May 10, 2018, from the Monash University website at https://www.monash.edu/rlo/graduate-research-writing/write-the-thesis/writing-the-thesis-chapters/reporting-and-discussing-your-findings

Data presentation and analysis. (2018). Retrieved May 10, 2018 from the Planning Tank website at https://planningtank.com/planning-techniques/data-presentation-and-analysis

Research data management system project: Best practices in research data management. (2018). New England Collaborative Data Management Curriculum. Retrieved May 10, 2018.

Rowley, J. (2002). Using case studies in research. Management Research News, 25(1), 16-27. Retrieved May 10, 2018. Available in the Trident Online Library. (Search by title and author’s last name.)

Stockberger, D. (2016). Introductory statistics: Concepts, models, and applications. Missouri State. Retrieved from http://www.psychstat.missouristate.edu/introbook/sbk19.htm

Brown, N., Lave, B., Romey, J., Schatz, M., & Shingledecker, M. (2018) Beginning Excel. OpenOregon, Creative Commons License. Retrieved from https://openoregon.pressbooks.pub/beginningexcel/ and https://openoregon.pressbooks.pub/beginningexcel/front-matter/introduction/

Book II: Chapters 1–4 and
Book V: Chapter 1 in:
Harvey, G. (2016). Excel 2016 All-in-One For Dummies. John Wiley & Sons. Available in the Trident Online Library: Follow these instructions for Finding Skillsoft Books. Enter 112925 in the search bar.

Video Material

ExcellsFun. (2012, October 19). Excel data analysis: Sort, filter, PivotTable, formulas (25 examples): HCC Professional Day 2012 [Video file]. Retrieved from https://www.youtube.com/watch?v=i5WiYh2jmG8

Gibbs, G. R. (2015, March 4). Quality in qualitative research [Video file]. Retrieved from https://www.youtube.com/watch?v=F1YfaSmDQbw

Curry, L. (2015, June 23). Fundamentals of qualitative research methods: Data analysis (module 5) [Video file]. Retrieved from https://www.youtube.com/watch?v=opp5tH4uD-w

Russell, D. (2014, October 24). Introduction to quantitative data analysis [Video file]. Retrieved from https://www.youtube.com/watch?v=k5XR3Ari7-0

McGinn, J., Kaniasty, E., Mistry, D., Soucy, K., & Snyder, C. (2012). Delivering results: How do you report user research findings? New Hampshire Usability Professional’s Association Meeting. Retrieved from https://www.slideshare.net/bobthomas/delivering-results-how-do-you-report-user-research-findings

Sampling and Data

Gibbs, G. R. (2012, October 24). Social surveys. Part 1 of 2 on surveys and sampling [Video file]. Retrieved from https://www.youtube.com/watch?v=M-lEVzKyqhQ&t=6s

Gibbs, G. R. (2012, October 24). Sampling. Part 2 of 2 on surveys and sampling [Video file]. Retrieved from https://www.youtube.com/watch?v=owN9hLq-Eac

Gibbs, G. R. (2014, March 11). Crosstabulations and their interpetation. Part 1 of 2 on crosstabulations and chi-square [Video file]. Retrieved from https://www.youtube.com/watch?v=B6bqHNVd-Kw

Gibbs, G. R. (2014, March 11). The chi-square statistic and reporting results. Part 2 of 2 on crosstabulations and chi-square [Video file]. Retrieved from https://www.youtube.com/watch?v=JmaL62bDsf8&t=5s

Privacy Policy | Contact

Best practices in researchdata management.pptx
Research Data Management System project:
Best Practices in Research Data Management*
*Adaptation of the NECDMC

NECDMC stands for New England Collaborative Data Management Curriculum, which is led by the Lamar Souter Library at Umass Medical School, in collaboration with several other NE libraries, including Tufts University. This presentation and discussion today utilize only a small amount of information available on the NECDMC website. The curriculum is designed to align with the NSF data management plan recommendations and to address some universal research data management questions and concerns.

This presentation is an adaptation of the NECDMC First Module, originally compiled and presented by my colleague Katie Houk, now working at Cal State San Diego.
1

Today’s Objectives
Why manage data?
Identify common data management issues
Best practices for managing data
Support: how the library and TTS can help you and your lab

So, this is a brief session that only scratches the surface of data management and data management planning issues concerns, but I hope that in about 20-25 minutes, we’ll gain a good understanding of these issues, understand the importance of data management, and have some new resources in your toolkit for approaching Data Management Planning. We will have an exercise following the lecture where you will break into groups and review a case study, and hopefully by the time I’m done with you today, you’ll have a firmer footing in this world, and you’ll know where to go for help. Spoiler, the Library is a great place to start…
2

What is Data?
“Research data, unlike other types of information, is collected, observed, or created, for purposes of analysis to produce original research results” (University of Edinburgh).
Observational
Experimental
Simulation data
Derived or compiled data 

There are a number of definitions for ‘research data,’ but this is a good one from the University of Edinburgh’s Data Management Handbook.

Data covers a broad range of types of information- generally categorized into broad areas such as Observational, Experimental, Simulation, and derived or compiled data.

Observational: data captured in real-time, usually irreplaceable. For example, sensor data, survey data, sample data, neurological images.
Experimental: data from lab equipment, often reproducible, but can be expensive. For example, gene sequences, chromatograms, toroid magnetic field data.
Simulation: data generated from test models where model and metadata are more important than output data. For example, climate models, economic models.
Derived or compiled: data is reproducible but expensive. For example, text and data mining, compiled database, 3D models.

Reference or canonical: a (static or organic) conglomeration or collection of smaller (peer-reviewed) datasets, most probably published and curated. For example, gene sequence databanks, chemical structures, or spatial data portals.

Can you think of any other types of data that get created during research? [ask about different types]

Documents (text, Word), spreadsheets
Laboratory notebooks, field notebooks, diaries
Questionnaires, transcripts, codebooks
Survey responses
Health indicators such as blood cell counts, vital signs
Audio and video recordings
Images, films
Protein or genetic sequences
3

Why Should I Manage it?
Transparency & Integrity
Compliance

You may be required by a funder or publisher to maintain the data that underlies your published works and findings.

Managing data is a part of compliance with University IRBs, and your funders’ data sharing and data management policies. Funders like the NIH reserve the right to audit your lab notebooks and pre-publication data; Since 2011 the NSF has required a data management plan and the federal govt. is currently working to make publicly funded research data available to the public.

The Fair Access to Science and Technology Research (FASTR) Act is a bipartisan effort aiming to make data from federally funded research more open and accessible. It was originally introduced in 2013 but not enacted, and was just reintroduced on March 18 2015. It follows up on the directive from the Office of Science and Technology issued in 2013 specifying that “The Administration is committed to ensuring that…the direct results of federally funded scientific research are made available to and useful for the public, industry, and the scientific community. Such results include peer-reviewed publications and digital data” (Holdren 2013).

Publications, private foundations and specific funders – like the American Heart Association – may also require data management provisions.

Essentially, the Age of Research Data has dawned. If you aren’t actively managing it now, you should be, because people are coming for it.

4

Science & Personal Benefits
Who uses your data now?
Who COULD use your data?
Shared/Open Data
Scientific progress
Impact on your career
Citation counts

Managing data saves you time and effort, and avoids the duplication of efforts, “good RDM = good research”. You can easily find the data you need and make these available should you be asked.

In addition, publishing your data can increase your citation impact and discoverability of your research & help with promotion and tenure.

You don’t know how someone else may use your data in the future.

Anna Gold. Cyberinfrastructure, Data, and Libraries, Part 1: A Cyberinfrastructure Primer for Librarians. D-Lib Magazine, September/October, 2007, Volume 13 Number 9/10 http://www.dlib.org/dlib/september07/gold/09gold-pt1.html.
“Managing and sharing data…
increases the impact and visibility of research;
promotes innovation and potential new data uses;
leads to new collaborations between data users and creators;
maximizes transparency and accountability;
enables scrutiny of research findings;
encourages improvement and validation of research methods;
reduces cost of duplicating data collection;
and provides important resources for education and training”
Increase the visibility of your research
Save time
Simplify your life
Preserve your data
Increase your research efficiency
Documentation
Meet grant requirements
Facilitate new discoveries
Support Open Access

5

What if I Don’t Consider RDM?

Data Sharing and Management Snafu in 3 Short Acts: A data management horror story by Karen Hanson, Alisa Surkis and Karen Yacobucci. 

This video from NYU lays a solid groundwork for the issues we will discuss today. In it are several scenarios that highlight data management issues that were identified by the Department of Health and Human Services’ Office of Research Integrity.

6

Seven “Issues” in Research Data Management
Responsibility
Data Management Plans
Records Management
File Management
File Naming
Metadata
Backup and Security
Ownership and Retention
Long Term Planning

Issue: Responsibility
Best Practices
Define roles and assign responsibilities for data management
Identify skills needed to perform tasks outlined in DMP and match to available staff
Develop training plans for continuity
Assign responsible parties and monitor results

Remember:
Data from federally-sponsored research belongs to the institution
You (Pis, grad students, lab workers) are stewards of the data
You need a plan to bridge the ever-changing staff in the lab
Your lab notebooks in any form may be audited by the funder.

Taking responsibility for your data, defining roles among your team, assigning responsibilities, and taking responsibility for training and continuity

Unless the distribution of responsibility is clear, misunderstandings can result and compliance jeopardized.

We hear a lot from students that they have had to learn DM on the go and may have little to no formal training on how to manage a specific project’s data, so do not be afraid to ask for clarification, and for documenting and formalizing DM roles & responsibilities. This is an important aspect of a DM plan.

8

Issue: Data Management Plans
Data Life Cycle

9

CREATING
DATA

PROCESSING
DATA

ANALYSING DATA

PRESERVING DATA

GIVING ACCESS TO DATA

RE-USING DATA

Creating a Data Management Plan
“the types of data, samples, physical collections, software, curriculum materials, and other materials to be produced in the course of the project;
the standards to be used for data and metadata format and content (where existing standards are absent or deemed inadequate, this should be documented along with any proposed solutions or remedies);
policies for access and sharing including provisions for appropriate protection of privacy, confidentiality, security, intellectual property, or other rights or requirements;
policies and provisions for re-use, re-distribution, and the production of derivatives; and
plans for archiving data, samples, and other research products, and for preservation of access to them”

This outline of a simplified Data Management Plan (DMP) is based on the NSF recommendations for its required 2-page data management plan

Some sections can be standardized based on your institutional practices

Remember when you craft your DMP as a requirement of funding that it is an executive summary for people who do not necessarily know much about your research or the process you will use. It is meant to be high-level and a broad overview. Going into too much detail could be counter-productive as it may make the document too long and might also make it too confusing for grant reviewers.

10

Issue: Data Management Plans
Best Practices
What types of data will be created?
Who will own, have access to, and be responsible for managing these data?
What equipment and methods will be used to capture and process data?
Where will data be stored during and after?

Many research funders require that you have a plan to manage and/or share your data. For example, in 2011 the NSF began requiring a data management 2-page supplement with all submitted grant applications. The NIH has requires a plan for projects in excess of $500,000.

These are some questions here are commonly addressed in a data management plan.

The NSF has laid the foundation for requiring a data management and sharing plan. You have a copy of a simplified data management plan. It is 7 sections with at least one question that should be answered per section to satisfy the requirements for a 2-page data management document.
11

Issue: File Management
Does this sound familiar?
Inconsistently labeled files
in multiple versions…
inside poorly structured folders…
stored on multiple media…
in multiple locations…
and in various formats…

If we think back to the video, we realize that a lot of the issues regarding data management relate to inconsistent and confusing file and folder labels, saving data in multiple locations, and not thinking about how someone might find and make sense of your data.

File management requires thinking about how you and others can both easily find and make sense of your data.

12

This slide comes from a NECDMC partner at Northeastern University. She looked at a sample of data files produced by students collecting data a bioscience lab.

As you can see, their file naming conventions do not always take into consideration how someone not involved in a project will make sense of what is in the file.

After some time, these files would probably not even make sense to the person involved in creating the file! I know we’ve all done this, and we all know it’s bad practice. If you can influence this sort of issue with your team or in your lab, you will already have conquered a great data management challenge.

13

Issue: File Naming
Best Practices
Avoid special characters in a file name.
Use capitals or underscores instead of periods or spaces.
Use 25 or fewer characters.
Use documented & standardized descriptive information about the project/experiment.
Use date format ISO 8601:YYYYMMDD.
Include a version number.

These are some best practices for creating file names. Poorly constructed file names can cause issues when transferring files from one format to another, or to another operating system.

For example, a researcher recently identified that when she moved files from REDCap™ to her analysis software, the dates were reformatted. In addition, OS like Unix can have issues reading files with spaces or special characters.

14

Issue: File Naming

Here is an example from a biomedical engineering lab that shows how you can add in project information into the file name. Notice that each file is with an experiment that links back to the laboratory notebook, so that there could be multiple people in the lab and multiple experiments involving the same sample, but having a systematic approach to labeling and mapping files allows for the efficient retrieval and interpretation of the data. There is no 100% right or wrong way to do this, but it is important to come up with a naming convention for files, images, experimental data, etc. before you start generating mass amounts of information. 30 minutes crafting a file-naming protocol will save hours of work and repetition down the line.

15

Issue: File Naming
Best Practices
Avoid special characters in a file name.
Use capitals or underscores instead of periods or spaces.
Use 25 or fewer characters.
Use documented & standardized descriptive information about the project/experiment.
Use date format ISO 8601:YYYYMMDD.
Include a version number.
Need Help?
Contact metadataservices@tufts.edu

These are some best practices for creating file names. Poorly constructed file names can cause issues when transferring files from one format to another, or to another operating system.

For example, a researcher recently identified that when she moved files from REDCap™ to her analysis software, the dates were reformatted. In addition, OS like Unix can have issues reading files with spaces or special characters.

16

Issue: Metadata
What is Metadata?
“Metadata is structured information that describes, explains, locates, or otherwise makes it easier to retrieve, use or manage an information resource.”
–2004, NISO, Understanding Metadata, pg. 1
A love note to the future…
How will someone make sense of your data e.g. the cells and values of your spreadsheet?
What universal or disciplinary standards could be used to label your data?
How can you describe a data set to make it discoverable?

Thanks to the NSA & the Edward Snowden scandal, ‘metadata’ has become a household word!

Often described as data about data, metadata are helps you structure and record information, to help you make sense of your data.

Metadata standards can be used to describe the data’s field labels, their values, elements and parameters, and they can also describe the nature of the files that are produced, such as how many bytes, the format, the software used to create the file, the version, and who created it.

17

Why Use Metadata?
find data from other researchers to support your research
use the data that you do find
help other professionals find and use data from your research
use your own data in the future when you may have forgotten details of the research
Help ensure consistency and clarity of data through the use of technical standards and controlled vocabularies

Metadata provides the ability to locate and use data from other researchers to help support your research

Helps others find and use your data

Facilitates the use of your own data long after you’ve completed the project when you may have forgotten all the details of your research

Metadata standards and controlled vocabularies ensures consistency of data by everyone on the project who is creating data.
18

Common metadata fields
Title
Creator
Identifier
Subject
Funders
Rights
Access information
Language
Dates
Location
Methodology
Data processing
Sources
List of file names
File Formats
File structure
Variable list
Code lists
Versions
Checksums

Here is a list of common metadata fields associated with a data set.
There are common elements necessary to ensure your data can be found and used by other researchers. They include:

Title – Name of the dataset or research project that produced it
Creator – Names of the organization or people who created the data
Subject – Best practice is to use a controlled vocabulary to establish the appropriate keywords or phrases
Funders – Organizations or agencies who funded the research
Access Information – Where and how your data can be accessed by other researchers
Methodology – How the data was generated, including equipment or software used, experimental protocol, other things one might include in a lab notebook
Data Processing – Along the way, record any information on how the data has been altered or processed
Sources – Citations to material for data derived from other sources, including details of where the source data is held and how it was accessed
List of file names – List of all data files associated with the project, with their names and file extensions (e.g. ‘NWPalaceTR.WRL’, ‘stone.mov’). Best practice is to establish a file naming convention to ensure ease of discoverability
Additional information is available on the documentation and metadata section of the Tisch Data Management research guide at: http://researchguides.library.tufts.edu/content.php?pid=167647&sid=1412592

19

Here is an example of metadata collected about a data set. Information includes author information, date the data was issued, doi identifiers, rights and embargo information, and a description of the data set.

20

What else?
Standard conventions are used to describe content in a way that ensures units such as date, time, location, etc. are entered consistently among the researchers in your group
Controlled vocabularies are lists of predefined terms that ensure consistency of use, and help disambiguate similar concepts. Use the controlled vocabulary that best matches your research.
You might create a short list of terms to choose from when populating a specific piece of data
For example, subject terms used in research about biometric sensing might be taken from a controlled vocabulary list such as Medical Subject Headings (MeSH)

Standard conventions – for example the ISO international standard for date and time is commonly used for expressing date and time related data. You might use standard latitude and longitude coordinates, or name codes to express location.
21

Issue: Metadata
Biology and health-specific metadata examples

Using REDCap you can upload or create a data dictionary to define the fields, elements, and parameters for your data collection.

Here is another example of metadata from a dataset uploaded to the NCBI “Flybase”.

It incorporates a large amount of scientific disciplinary information such as the strain, tissue, and cell line used in the sample.

Most databases where you upload your data will inform you of the basic metadata they require. Many data repositories actually have experts that create more metadata after you submit in order to make it findable and interpretable.

22

Issue: Metadata
Best Practices – Create a Data Dictionary
Describe the contents of data files
Define the parameters and the units on the parameter
Explain the formats for dates, time, geographic coordinates, and other parameters
Define any coded values
Describe quality flags or qualifying values
Define missing values

Need Help?
Contact metadataservices@tufts.edu

A data dictionary is a “centralized repository of information about data such as meaning, origin, relationships to other data, usage, and format.” (wikipedia)

Use a data dictionary to define the elements, technical standards, and controlled vocabularies you use in your project.

Include information that describes the data set as a whole as well as the field specific metadata that provides detail specific information about your data.

23

Metadata and the ELN
Any searchable field in the Agilent or LabArchives ELN technically contains metadata
In both ELNs, you can add tags/keywords to experiments, data files, and image files
In some cases you can create a pre-defined list of tags/keywords to choose from

Neither vendor in this pilot employ metadata standards for your use BUT

You should create a data dictionary of the elements, technical standards and controlled vocabularies you use in your project.

In the ELN use tags and keywords generated on the fly help locate experiments, files, and other data in the ELN. But beware of misspellings: “tae” not “tea”

Pre-defined lists relies on controlled vocabulary, thus reducing the chance of human error

24

Agilent
Searchable fields:

Here’s an screen shot that illustrates searchable fields in the ELN
25

Agilent
Funding Source via menu:

Project Focus via menu:

26

Agilent
Associate metadata with an experiment using keywords:

Keywords help with searching, but keyword metadata is not sufficient for making your data accessible and reusable by resarchers.
27

LabArchives
Associate metadata with an experiment using tags:

Just like in Agilent, in LabArchives you can apply tags or keywords to experiments.
It’s easy to do. Tags/keywords are visible and searchable immediately upon creation.
28

LabArchives
Associate keyword metadata with an image file:

29

Issue: Backup & Security
How often should data be backed up?
How many copies of data should you have?
Where can you store your data?
How much server space can I get?

IRB guidelines and IT departments can help you learn where and how to best store & backup your data
30

Issue: Backup & Security
Best Practices
Make 3 copies (original + external/local + external/remote)
Have them geographically distributed (local vs. remote)
Use a Hard drive (e.g. Vista backup, Mac Timeline, UNIX rsync) or Tape backup system
Cloud Storage – some examples of private sector storage resources include: (Amazon S3, Elephant Drive, Jungle Disk, Mozy, Carbonite)
Unencrypted is ideal for storing your data because it will make it most easily read by you and others in the future…but if you do need to encrypt your data because of human subjects then:
Keep passwords and keys on paper (2 copies), and in a PGP (pretty good privacy) encrypted digital file
Uncompressed is also ideal for storage, but if you need to do so to conserve space, limit compression to your 3rd backup copy

Electronic data should be saved on a device that has the appropriate security safeguards such as unique identification of authorized users, password protection, encryption, automated operating system patch (bug fix), anti-virus controls, firewall configuration, and scheduled and automatic backups to protect against data loss or theft.

31

Issue: Ownership & Retention
How long is long enough?

“How long should I retain data?” is not a clear and cut data management question. Last June, for example, the Journal of Clinical Investigation retracted a published article after 6 years because one of its data tables was duplicated.

The publisher contacted the researchers to have them update the data, but they could not locate the original data files after six years, so the journal was forced to issue a retraction.

This case highlights how difficult it is to know for how long to keep data. This article was peer reviewed and cited over 55 times but it took six years for the representation of its data to be called into questioned. Thus, thinking about ways to digitize documents and store and preserve electronic files of data in a self-archived, disciplinary, or local data repository is important, and one of the many tasks the library can help you with.

RetractionWatch.com has three different categories dealing with retractions due to various data misconduct. Fabrication of data, duplication of data, manipulation of figures/images. There’s also the issue of non-reproducible results. All of these issues can be avoided by honest researchers by good data management and preservation practices.

32

Issue: Ownership & Retention
Intellectual Property Policy
IRB data retention policy
Funders’ data retention policy
Publishers’ data retention policy
Federal and State laws

When it comes to data ownership and data retention there are a lot of overlapping policies. IP policy can cover the ownership and retention of data related to patents, the IRB wants to ensure that documentation of human subjects’ data are retained and/or destroyed appropriately, and the funders and publishers want you to retain data to defend the integrity of your findings, and then there are federal guidelines like HIPAA.

33

Issue: Long-Term Planning
What will happen to my data after my project ends?
How can I appraise the value of my data?
What are my options for archiving and preserving my data?
What are my options for publishing and sharing data?

After a project you may want to consider appraising, and publishing or depositing your data in a repository. There are a variety of factors that impact your ability to share data with outside parties. According to the OHRP, you should contact the IRB prior to proceeding with a release of human subject data unless (a) your subjects signed an IRB approved consent document with HIPAA compliant authorization language that clearly details what information will be collected, used, and disclosed and (b) the outside party is specified in the document.

Archiving & Preserving versus Storage – there’s a big difference! Digital data degrades if it is not properly taken care of. Depositing data in a repository for preservation and open access ensures that data will be properly cared for throughout the rest of it’s life – however long you determine that to be. If it is forever, then the repository will migrate the data onto the newest, most abundant storage media and convert it into a format that can be interpreted by computers in the future. (Think of the 3.5in floppy, the zip drive, the cassette, etc.)
34

Open vs. Proprietary Formats Used in Research Labs

This graphic was created by a colleague that observed the number of instruments in just one biomedical lab relying on proprietary software.

This means that to be able to open and view this file, someone would need to know the software that created it, and be able to access that software. Thus converting your files to open source and sustainable formats and standards are essential for long-term sharing, preservation and access.

35

Issue: Long-Term Planning
Best Practices
When choosing a file format, select a consistent format that can be read well into the future and is independent of changes in applications.
Non-proprietary: Open, documented standard, Unencrypted, Uncompressed, ASCII formatted files will be readable into the future.

Works Cited
Lamar Soutter Library, University of Massachusetts Medical School. 2014. “New England Collaborative Data Management Curriculum: Module 1.” http://library.umassmed.edu/necdmc.

DataONE. 2013. “Best Practices for Data Management.”
http://www.dataone.org/best-practices.

MIT Libraries. 2013. “Data Management and Publishing.” MIT
http://libraries.mit.edu/guides/subjects/data-management/index.html.

Office of Research Integrity. 2013. “Data Management.” United States Department of Health and Human Services. United States Federal Government.
http://ori.hhs.gov/education/products/rcradmin/topics/data/open.shtml.

This work is licensed under a Creative Commons Attribution
– NonCommercial
-ShareAlike 3.0 United States License.

37

Learn More
Data Management Principles & Education:

Tufts Libraries Data Management Guide

Research Data MANTRA

DataONE: Best Practices

UK Data Archives

MIT Data Management and Publishing Guide

Data Management Plans

Digital Curation Centre

DMPTool2

DataONE: Data Management Planning

38

Find Help
Data Management Plans and Metadata services:
Medford/Somerville Campus: [names/contact info]
Boston/Grafton Campus: [librarian names/contact info]

Data storage and security services + ELN support:
[TTS contact info]

39

Finding Skillsoft Books

Finding Skillsoft Books in

Additional Library Resources

1. On the Portal page, click the link for Additional Library Resources.

2. Scan the list of available databases until you find the link for Skillsoft Books (BusinessPro and
ITPro). Click on the link.

3. Enter your search item in the Search box and click GO (or press Enter on your keyboard).

4. Your search results appear. Click on a book title to open that book.

5. Click on links to access chapters and other book content.

Modules/Module5/Mod5Case.html

Module 5 – Case

Comparing Models to Assess the Research Problem

Assignment Overview

In this Case Assignment, you will be reading several articles and sources relating to different kinds of data analysis, the process of inferring conclusions from data, the legitimacy of different kinds of analyses, and the development of conclusions and recommendations from your data. You will then be asked to develop a preliminary data analysis and presentation plan, describing the kinds of inferences you hope to draw from your data and how you justify their legitimacy. You also be asked to sketch out a plan for presenting your findings for your research site and wrapping up your relationship with the site.

As the course has progressed, you have made a series of decisions in which you have increasingly specified the structure of your project. These decisions build on each other in critical ways. However, you’re not locked into previous decisions if you subsequently decide you need to change direction. As before, you have an opportunity to go back and revise and/or extend the sections completed in previous modules. If you do revise earlier sections, please include them and indicate what changes you have made to them. Case grades given for the first four modules have been advisory rather than final. Your ultimate grade will be based on the completed methodology section represented by the final versions of all five cases put together at the end of the course here in Module 5.

Case Assignment

Prepare a 5- to 7-page paper in accordance with the following Assignment Expectations, describing the analysis issues for your project that form the basis for the Methodology section of your dissertation, in accordance with the following outline:

Identify the major kinds of data that you plan to collect during your project. Discuss the IRB approval process. For each major kind of data, identify steps that you will take to analyze those data for purposes of the project. The readings will help you identify different kinds of data and appropriate analytical procedures associated with each.
Identify the main kinds of conclusions that you hope to draw from your data analysis. These might be descriptive/analytical, prescriptive in the form of recommendations, suggestions in the way of forward thinking, or some combination of these. For each type of conclusion, identify offices and people within the organization that you believe might be interested in your conclusions, and why they might be interested.
Identify the main deliverables from your project, and to whom the delivery would be made other than the members of your doctoral committee.

Remember, the five Case Assignments that you’ve completed in this course together constitute an effective first draft of the Methodology section of your project proposal. Therefore, you want to review the previous assignments carefully in light of the feedback that you have been given and take the opportunity to make any needed revisions or modifications so that the document hangs together clearly and effectively. Your final grade for the case part of this course will be based on the sum of the five parts taken together, so this is your opportunity to remediate any deficiencies you might have encountered in previous modules.

Please conclude your paper with a paragraph or two assessing the effectiveness of this approach to developing the methodology for your project and any lessons that you have learned during this course about the nature and conduct of academic research in applied settings. 

Assignment Expectations

Length: The written component of this assignment should be 5–7 pages long (double-spaced) without counting the cover page and reference page.

Organization: Subheadings should be used to organize your paper according to the questions.

Grammar and Spelling: While no points are deducted for minor errors, assignments are expected to adhere to standard guidelines of grammar, spelling, punctuation, and sentence syntax. Points may be deducted if grammar and spelling impact clarity. We encourage you to use tools such as grammarly.com and proofread your paper before submission.

When you write your paper make sure you do the following:

Answer the assignment questions directly.
Stay focused on the precise assignment questions. Do not go off on tangents or devote a lot of space to summarizing general background materials.
Use evidence from your readings to justify your conclusions.
Be sure to cite at least five credible resources.
Make sure to reference your sources of information with both a bibliography and in-text citations. See the Student Guide to Writing a High-Quality Academic Paper, including pages 11-14 on in-text citations. Another resource is the “Writing Style Guide,” which is found under “My Resources” in the TLC Portal.

Your assignment will be graded using the following criteria:

Assignment-Driven Criteria: Student demonstrates mastery covering all key elements of the assignment.

Critical Thinking/Application to Professional Practice: Student demonstrates mastery conceptualizing the problem, and analyzing information. Conclusions are logically presented and applied to professional practice in an exceptional manner.

Business Writing and Quality of References: Student demonstrates mastery and proficiency in written communication and use of appropriate and relevant literature at the doctoral level.

Citing Sources: Student demonstrates mastery applying APA formatting standards to both in-text citations and the reference list.

Professionalism and Timeliness: Assignments are submitted on time. 

Privacy Policy | Contact

Modules/Module5/Mod5SLP.html

Module 5 – SLP

Comparing Models to Assess the Research Problem

In this SLP you will use statistical tests and graphics to determine if measures are significantly different across some specified criteria. Since your research is focused on one firm, you will be testing across relatively small sets of data. With such small sets of data, the tests do not assume normality that is required of parametric statistics. In parametric statistics, samples are drawn from a population so as to make generalizations from that population. For the tests you are doing in your dissertation, you will use small sample t-tests and visualizations.

For the United Way, this spreadsheet shows the donations across 14 of their major corporate partners. The donations are broken out by gender. Not only are they interested in the total gifts from each firm but also the break-out by gender and the average per person donation. Here is an image of the clickable spreadsheet shown in Figure 7.

Figure 9. Pivot Table example

SLP Assignment Expectations

Using the secondary data for your selected firm, produce an analysis that compares some key criteria to show the selected areas of focus in your research on your firm. This analysis will help you to show the importance of your research question for this firm and set the stage for the qualitative research you are about to undertake.

Produce a spreadsheet with associated graphs, also provide a page or two to discuss this data analysis and the conclusions you have drawn. Add this to the growing work you have on the Background for the firm you are studying. By integrating this work into your Background section, you will have completed an important addition for the appendix of your Dissertation. So be sure that you put together all five SLPs that tell the story of why you selected this firm for your Dissertation and how your analysis supports the importance of the work you will be doing on this dissertation research. 

Your assignment will be graded using the following criteria:

Assignment-Driven Criteria: Student demonstrates mastery covering all key elements of the assignment.

Critical Thinking/Application to Professional Practice: Student demonstrates mastery conceptualizing the problem and analyzing information. Conclusions are logically presented and applied to professional practice in an exceptional manner.

Business Writing and Quality of References: Student demonstrates mastery and proficiency in written communication and use of appropriate and relevant literature at the doctoral level.

Citing Sources: Student demonstrates mastery applying APA formatting standards to both in-text citations and the reference list.

Professionalism and Timeliness: Assignments are submitted on time.

Privacy Policy | Contact

Figure 7

module5tests.xlsx
profitnonp

Row Labels Sum of 2015 Donations

NonProfit 3555041

Profit 10196119

Grand Total 13751160

MenvsWomen

t-Test: Paired Two Sample for Means For Men (Variable1) to Women (Variable2)

Variable 1 Variable 2

Mean 671584.583333333 474345.416666667

Variance 762499896487.901 463181017437.72

Observations 12 12

Pearson Correlation 0.9660030215

Hypothesized Mean Difference 0

df 11

t Stat 2.4540547774

P(T<=t) one-tail 0.0160073422 one tailed men were signficantly higher givers at p<.02 t Critical one-tail 1.7958848187 P(T<=t) two-tail 0.0320146845 two-tailed Men were significanatly higher givers at p<.03 t Critical two-tail 2.2009851601 pivot United Way Corporate Partners Sample Company Number of Employees Type Male donations Female Donations 2015 Donations Per Person Donation University of Ca, Irvine 23605 NonProfit 896530 534870 1431400 60.63969498 St. Joseph Health 11925 NonProfit 326890 226582 553472 46.41274633 California Stat Univ, Fullerton 5781 NonProfit 458239 268498 726737 125.7112956 Hoag Memorial Hospital 5240 NonProfit 98643 458930 557573 106.4070611 AAA of California 3600 NonProfit 187402 98457 285859 79.40527778 Walt Disney 29000 Profit 3278976 2578333 5857309 201.9761724 Allied Universal 8229 Profit 89765 54300 144065 17.50698748 Kaiser Permanente 7694 Profit 187562 176300 363862 47.29165584 The Boeing Company 6103 Profit 765320 325890 1091210 178.7989513 UnitedHealth Group 3900 Profit 654890 489236 1144126 293.365641 Tenet Healthcare Corp 4346 Profit 238456 134876 373332 85.90243902 The Irvine Company 4200 Profit 876342 345873 1222215 291.0035714 Determine if there is a signifcant difference of donations across For Profit and Non-Profit Organizations Per Person Donation University of Ca, Irvine St. Joseph Health California Stat Univ, Fullerton Hoag Memorial Hospital AA A of California Walt Disney Allied Universal Kaiser Permanente The Boeing Company UnitedHealth Group Tenet Healthcare Corp The Irvine Company 60.639694980000002 46.412746329999997 125.7112956 106.40706110000001 79.405277780000006 201.9761724 17.506987479999999 47.291655839999997 178.7989513 293.36564099999998 85.902439020000003 291.0035714 Modules/Module5/Mod5Objectives.html Module 5 - Outcomes Comparing Models to Assess the Research Problem Module Create basic data analytics models to predict and prescribe decisions and actions for complex business problems. Case Assemble the methodology chapter. SLP Use data and visualization tools to complete the background. Discussion Reflect on the dissertation task ahead. Privacy Policy | Contact Table of Contents.html   DOC670 Applied Statistics for Research (WIN2021-1) - Module 5: Comparing Models to Assess the Research Problem 1. Home 2. Background 3. Case 4. SLP 5. Learning Outcomes