Scientific Method

We are already familiar with the scientific method and that provides background and context for what scientists do. It may be helpful to briefly review that process so we see that the scientist begins with a problem, problem questions, and ends with widely shared results and discussion. Note that the academic reward system has little interest in research as such. Rather the key is publication and the visibility/reputation resulting from that publication. Scientists working for government and commercial agencies may be limited in what research results may be disclosed because of security concerns or protecting intellectual property.

If you were asked, what is the single most important characteristics of the scientific method?

Education

While science may be done without graduate education [many scientists were amateurs before the 19th and 20th Centuries], most scientists will have graduate degrees. Those with MS degrees are often limited to assistant or specialized roles in the laboratory or with instruments and measurement. In applied research, those with considerable field or on-the-job experience usually have graduate education, but not always. There is still some room for bright amateurs [the basement inventor]. The Ph.D. degree is the terminal degree for those who wish to practice research in an academic setting and in many others as well. Often, those with the Ph.D. with have two to three undergraduate years devoted to scientific study and three or more graduate years. The purpose of the Ph.D. program is to train researchers and the doctoral dissertation [which should be publishable] is the cumulating product of many years of education. Many Ph.D. students will work in the laboratory of an established researcher as a research assistant or research associate. They are immediately involved in assisting a research project, normally a funded one [funded by an external agency]. This on-the-job experience provides first-hand experience in the nuts and bolts of the research enterprise, including fund-raising and laboratory management. Ideally, the faculty member in charge of the research [the principal investigator] will introduce the doctoral students at professional meetings and help them to network with others in the field. At the same time, students learn from and help each other. Typically, since science papers [research articles are multi-authored], the student will be one of several authors on research papers published on the research being done in the laboratory [but not as the first or second author].

The dissertation-related publications, post-doctoral experiences [in other labs under other PIs], presentations at scholarly meetings, and recommendations from well-regarded PIs help the new Ph.D. to find an appropriate position. Still, the job market can be tight, especially for those interested in pure science, and it's not unusual to have difficulty in finding a position. What has been called "the invisible college," the network of those who conduct research and publishing in a particular field and communicate via the Internet, telephone, and face-to-face meetings at conferences, plays a major role in placement. The better the reputation of the University, of the Department, of the academic advisor, and of the PI in the laboratory, the more likely that a good position will be found.

If you were hiring a new STM Ph.D., what attributes would you look for? How important is the university where the degree was earned?

The Lab

While circumstances may differ by discipline and problem area, the scientist needs a laboratory to conduct her research. The lab may be an institutional one or it may in effect belong to a particular principal investigator. In addition, field work may also be needed to observe, measure and gather data. Lab space and equipment may hinder or facilitate research. Typically, it is difficult to find enough space and to gain access to the best equipment. Our new Ph.D. needs to become, as quickly as possible, a principal investigator [the person in charge of and responsible for the research project]. This means that she must have a research project or more than one ready to go [at least conceptually]. Often, this will be related to the doctoral research and/or research done while a post-doc in another laboratory.

In a government or commercial setting, funding is normally proved by the agency [internal funding]. However, the scientist is still responsible for creating and maintaining a persuasive argument to justify adequate funding. If he cannot convince those who make budget decisions that his research is important and necessary, support is likely to be limited or he may find himself working on research that reflects another's priorities rather than his own.

In an academic setting, new faculty are normally given "start-up" funding to insure that adequate space, equipment, and staff are available for research to begin. However, this funding is often available for the short-term on the assumption that the researcher will find external funding after that. Thus, one of the key elements in the practice of science is the ability of the scientist, working with others at his place of work, to find the funding needed to begin or continue research [the scientist as fund-raiser]. Since much research funding is short-term, i.e. three or four years, fund-raising including persuasive, positive reports to funders is a continuing aspect of what the scientist does.

While pure science, pushing back the forces of ignorance, may be an ideal, the real world increasingly involves the ability of the scientist and her institution to match the research interests of others, usually government agencies or commercial firms. This means that the research agenda is external rather than internal.

Why might this be a major problem for the scientist and the university?

The Environment

While the scientist may wish to be in the field or in the lab with his or her colleagues, science is very much a social and political activity where public and private interests collide. For example, how might externally funded drug research create problems for the researcher?
The reaction of the Bush Administration executive and legislative branches to global warming research done by government scientists is a good example. Another is that externally funded research may have strings or conditions that prohibit disclosure or full disclosure. Too, research funded by a corporation with a product to sell may yield "corrected" or "adjusted" results when published.

Can you identify environmental pressures most likely to impact the scientist and her research?

Principal Investigator [project director]

When funding is made available, either internally or externally, the scientist has become a PI or project director. This means that besides being the lead scientist,  administration will involve a substantial amount of time, time that most scientists would rather spend in the lab doing science. In general, the PI is responsible for:

• Creating, editing, and submitting the research proposal, including the budget and the time line to completion
• Insuring that the project is completed according to deadline
• Selecting, evaluating,  and supervising project staff
• Managing the budget and able to account for all project expenditures [includes rebudgeting if needed]
  
• Preparing and submitting [revising as needed] all required forms for human subjects, the use of live animals, hazardous materials, and the like
• Insuring that no problems exist with subjects or materials used and secures remedies if problems arise
• Negotiating with funding agency [sponsor] as needed
• Insuring that data collection meets appropriate standards
• Insuring data integrity and preservation [six years for federal funding]
• Preparing and submitting monthly and other reports to university and funder
• Certifying the effort of those involved in the project
• Maintaining adequate and accurate records.

Thus, the lead scientist spends substantial time and effort on gaining and retaining funding as well as project management and supervision.

Would you like to be a PI? Why?

Straight Line Science

Given the portrait of the scientist often seen in popular media, laypeople often consider science to be a straight line process where the scientist begins at A and soon ends at Z. In fact, things often go wrong. Equipment doesn't work properly. Staff make errors. The predicted relationship is not generalizable. The sample is too small or atypical. Results are interesting, but unexpected. Referees [subject experts who evaluate articles for publication] reject the findings as trivial or the conclusions as unwarranted. Acknowledged experts in the field of study are often closely tied to conventional wisdom and/or traditional research techniques. Unexpected results may be rejected.

The scientist must not only do the science well, but she must also be credible -- be able to convince others that the science was properly done and that the results are valid [measured what was intended to be measured] and reliable [others would find the same result]. Likely critical comment must be anticipated and rationale created to counter it. Research will be of little value if the researcher cannot convince others, the editorial board members who are referees, readers of scholarly journals, and funding agencies that the research is important and that the results are meaningful. Research results do not always stand on their own. They require persuasive advocacy.

Does it seem reasonable that science might move forward with stops, starts, and surprises?

Teaming

Research in the humanities is usually a one person operation. In the social sciences, most research is done by a single individual, but multi-person research is increasing. In science, perhaps because of the cost of equipment, space, and the like, nearly all research is done by a team under the leadership of the principal investigator. This means that scientific research will normally have multiple authors and individuals may have specialized roles with one person focusing on instruments and measurement, another on data analysis, and so forth.

The cost of team research results in the need for substantial funding and the emphasis on external funding since an individual would lack the resources and the research to do scientific research. Even in technology research, the basement inventor is far outnumbered by the large corporate and government laboratories.

Why would team research be substantially different than individual research?

The Sad Truth

Although discouraging, the evidence is clear. Most scientists seek information from colleagues rather than from the library or the information center. Both locally and beyond, the "invisible college" serves as an information network on all aspects of the scientific enterprise. Ideally, the STM information professional would be part of that network. It is difficult, but one way to begin to build bridges is to spend time with scientists in their place and to attend the scholarly meetings that they attend. Working with graduate students in effective and innovative ways, with the new scientist when she arrives at your place, will create awareness and help to develop awareness of the various ways that we can help the scientist to be more successful.

While there are exceptions, we often deal with proxies rather than the scientist. This may create a variety of problems since the information wants/needs may not be clearly shared with the graduate student, secretary, administrative assistance who is charged with getting desired items from the library. While the Internet and patron-generated ILL requests, make it relatively easy to request items and have them delivered electronically or via campus or plant mail, it does make it more difficult to have reference interview type situations.

What might the individual or the agency do to make the library or the information center more visible to the scientist?

Current Awareness

The successful scientist must be aware of current developments in his discipline and especially in his research area. The scientist will likely have a research agenda, both for the immediate future and a longer-range one for some day. Much of the awareness will be gained from society periodicals, websites, RSS feeds as well as informal communication via discussion lists and email. Since there is considerable urgency to completing research before another, awareness of new developments, of who is doing what and why is absolutely essential. Scientists typically need no help with current awareness unless they are beginning interdisciplinary research and need to come up to speed with a new content area. New scientists, because of the cost of memberships and subscriptions, are more likely to use current periodical collections at the library.

Since funding, especially external, is the foundation of research, scientists are especially concerned with news of new grant opportunities, both from government and commercial sources [and sometimes from foundations]. Most universities and larger corporations have a research support organization that subscribes to funding alert services. Sometimes, this will be done by the library or the information center. A good example of a basic source is http://www.grants.gov/. The University of Illinois at Urbana-Champaign is a good example of an aggregated service. It is essential that you be familiar with the services available where you will be informed. This is  a good topic for an instructional session for STM graduate students.

How to you learn about new research findings in STM?

The Research Agenda

Each scientist has some sort of research agenda. This may change over time as funding and departmental priorities change. Knowing the research agenda of each scientist, as well as those of doctoral students, is the building block of serving STM professionals. With this knowledge, you can create a STI [systematic dissemination of information] service formally or informally and perhaps even become part of the research team. Note that most funding agencies have research agendas and that these change from time to time. Ideally, you would be aware of the research agendas of the the leading funding agencies for those that you support [NSF for example]. Here is an example from the Centers for Disease Control and Prevention.

Would information professionals have a research agenda? Would the agency have one?

Researching--Getting Started

Most research will continue with the topics and methods used in the dissertation and the post-doc experience. The scientist has a research agenda when she received the degree and finds employment. That agenda may be modified by the employer's needs. Depending on the freedom allowed, the scientist will select a problem area and create problem questions or hypotheses to provide the framework for data collection, measurement, and analysis.

Typically, the first stage of research is the literature search. This should involve the library or the information center in providing identified items or helping to identify likely items. Most scientists use a pearl-fishing approach. That is, the find a good recent article and work backward in the literature via references or citations until no more useful literature is found. The key is that first article and the quality and comprehensiveness of the literature review. Indexing and abstracting services are less popular. Web of Science, if available is a popular choice, especially because it allows reference searching. The literature review process involves:

1. Finding relevant literature
2. Evaluating the utility and importance of the literature
3. Preparing a review of the relevant literature that clearly indicates gaps and strengths
4. Indicating how the planned research will build upon what is known and fill a gap that needs to be filled.
   

Researching--Data Collection

Data collection, including sample size, will vary depending on the science or technology and whether the work is done in a lab or in the field. Depending on the situation, this may take considerable time and effort. The process is easier if the scientist can use methods demonstrated to be successful in a previous study. The data collection process involves:

1. Insuring that the variables and values are appropriate and are operationally defined
2. Collecting enough data so that the results will be generalizable
3. Insuring that the data is valid and reliable
4. Capturing the data in a form that allows relatively easy analysis
5. Insuring that the data is preserved.

What role might a library or information center have in providing access to and preserving research data?

Researching--Data Analysis

When the data is collected, it must be analyzed. This is normally done using a statistical software package such as SAS or SPSS. There are also specialized statistical software packages for specific disciplines and discipline clusters. The data analysis process involves:

1. Selecting the appropriate statistical software
2. Understanding which statistical tests are relevant/appropriate
3. Running the statistical tests
4. Capturing the results and understanding what they mean
5. Creating tables, graphs, figures or text summaries that clearly indicate what was found.

Have you used statistical software?

Researching--Results

With the data analysis summaries in hand, the scientist will carefully list major and minor results. Since periodical articles are limited in the number of words and pages, only the most notable results will be reported. In a dissertation or technical report, it may be possible and appropriate to report all of the results. A paragraph or so will be prepared for each major finding or result. After the results are clear and persuasively stated, the results will need to be compared to relevant findings in the literature to note where there is agreement, disagreement, and new findings.

Researching--write up

Since the purpose of research is publication and sharing, target publication outlets were identified before the research began. In some cases, a query letter was sent to ascertain interest. Using instructions to the author and a recent article as a template, the scientist will prepare the article. After draft-sharing, a preliminary version of the paper may be shared at a conference or as a preprint. The paper will then be sent to a periodical where it will be reviewed by the editor and, if seen as worthy, sent to subject experts for their evaluation. If they agree that the paper is publishable, it will normally be returned to the author to make suggested revisions. When revised, the paper is scheduled for publication which may be several months away. If the paper is rejected, the author makes changes as appropriate and submits to a second periodical.

While this is going on, the author is working on the next research project which may be an extension of this one.

Teaching

Scientists affiliated with a university or college are normally required to teach although they may, with enough external funding, buy out their teaching and not teach at all except for informal instruction in their laboratory. While courses taught would ordinarily be in the same areas where the research is being done, research is deep and narrow while courses are much broader, but also shallow. Preparing course material, evaluating student work, and being available for students takes some time and effort. The ideal, not always realized, is that research informs teaching and vice versa.

The ideal is that teaching informs research and that research informs teaching. Does this seem reasonable?

Service

Whether in academic, government, or commercial environments, the scientist will have some service responsibilities. Typically, this means serving on committees dealing with various topics. Some organizations require considerable meetings to discuss policies, procedures, plans and assessment. Again, this takes time away from research. Still, internal networking is important politically.

Another aspect of service is related to being a member of appropriate professional or scholarly societies. Again, serving on committees such as the editorial board of the society periodical or being involved in conference planning [nearly every society has a major annual meeting. Many scientists are members of and are involved with several societies.

Summary

When all is done, it is the research publication that defines the scientist in terms of his or her discoveries and contribution to what we know about the natural world. For example, the number of citations for a scientist's work is often seen as validation of that person's value and worth. For the engineer or applied scientist, patents and the value of intellectual property created loom large in creating a reputation.