|
| |
Research Question and Annotated Bibliography
Formulate a researchable question, and locate at least five published reports
of original research to help answer it. Sometimes it's easier to find at least
some of the research first and then formulate the question. Even if you
formulate a question first, you may reformulate it once you've started reading
the existing research.
- Identify and select a relevant topic that has yielded a significant body of
related research
- Locate and select appropriate research studies
- Define a research question or hypothesis around which you can organize
your paper
You can use this assignment to get a start on the
research paper; see the description of that assignment
for ideas about some possible topics. Some ways to locate research studies you can use:
- Check databases like ERIC, Library Literature, and EBSCO. Limit
your search to peer-reviewed journals. Use the names of research methods as
descriptors.
- Go to the library and browse through peer-reviewed journals in your area.
Look for research articles that appeal to you. When you find them, use their
bibliographies and/or Social Science Citation Index to find others
that are related.
- If you pick Research Paper Option 2 ("Evaluate the
research of a productive LIS scholar"), start by looking for five or
more articles on the same general topic by that scholar -- and/or by others
who cite that scholar's work. Check the "more research needed"
comments in the conclusions of the earliest articles. Do they forecast the
scholar's subsequent work? Maybe this will help you define a research
question or hypothesis, too.
A good reason to hand this assignment in on time, even if you don't think
it's ready: feedback on this assignment can improve your research paper. Your
grade on this assignment may be raised retroactively if you make good use of
feedback.
When you hand in this assignment, it should consist of:
- A first-draft introduction to your literature review, including a
problem statement (specifying the "so what?" of your investigation –
why should we care about the answers?) and a purpose statement (specifying your
exact question or hypothesis, in a researchable form). The problem statement should place the literature review firmly in either a
research context, a professional context, or both. Examples:
 |
Research context: "Previous studies of reference question
formation had assumed that questioners are motivated by their own felt
need for information. When Gross (date) identified the 'imposed query'
-- for which the questioner's motivation may be a school assignment, a
work supervisor's request, or a friend's need -- reference librarians
immediately recognized the phenomenon. This paper will examine how the
work of Gross, Saxton, and others has integrated the 'imposed query'
phenomenon into existing models of information seeking behavior." |
 |
Professional context: "A decade ago, as the Internet gained
visibility, the famous 'digital divide' between information haves and
have-nots was put forth as a major reason to fund technology in public libraries.
Since then many studies have shown that children who have home access to
technology perform better on tasks X, Y, and Z (citations go here). The
issue has not gone away, and in light of the No Child Left Behind Act it
seems more important than ever." |
- Purpose statement should state one or more clear research questions or
hypotheses, defining or predicting the focus and structure of the paper.
Notice that good hypotheses and research questions contain variables, which
will need to be defined. Examples:
 | Research
questions: "How successful have libraries been in
helping to narrow the digital divide? To help answer the question, we
will look for evidence of
-
what funding the libraries have won for
their technology;
-
what level of Internet access public libraries now
provide; and
-
how public library Internet access is used."
|
 | "Success in narrowing the digital divide" is a
variable, but you need to define it better before you can measure it. In
the long range, "success" might mean "making sure that
kids without home computers can succeed in school and work at the same
rate as kids who do have home computers." But to measure that would
need a longitudinal study -- it could take years -- and it would be
difficult to screen out other influences on school and work achievement
-- do kids with home computers also have better educated, richer
parents? So you might want to look at "operational
definitions" of "success" that are more like the
shorter-range objectives in your planning process. The first -- securing
funds -- would be relatively easy to measure: how much money? how many
grants? The second is also fairly easy -- how many Internet access
points per member of the population served? The third looks harder, and
you might need to operationalize some more. How many libraries have time
limits on terminals? How many people are waiting to use terminals at any
given times? What do they use them for? Can you find out without
violating their privacy? |
 | Hypothesis: "There should be a significant relationship between
the librarian's use of probe questions and the successful outcome of the
reference transaction."
-
Independent variable:
The librarian's use of probe
questions. To measure this, you would have to define "probe
questions." Would you just want to count all probe questions,
or would you sort them into types of probe and count each
type?
-
Dependent
variable: Outcome of the reference transaction. How would you
define "success"? Right answer? Relevant retrievals, with
not too many irrelevant ones? Customer satisfaction?
|
 | You
would probably find that different studies you reviewed would have
slightly different operational definitions for each of these variables,
and you should address the differences in your paper.) |
- An annotated list
of the research reports. (You may also find other materials, such as published
literature reviews summarizing the existing research, or relevant opinion
pieces. You may include these in your bibliography but annotating them is
optional. Consider them supplementary to the reports of original research.) Each
annotation should include a complete bibliographic citation, and should
summarize briefly 1) what the researcher(s) intended to find out (or prove); 2)
what methods were used (e.g., survey, participant observation, systems
analysis); 3) strengths and/or weaknesses you can discern in the way the study
was conducted; and 4) what was in fact discovered -- the findings.
Your introduction and annotations will be written in complete sentences, but
need not be long or detailed; this assignment should take 2-3 pages. Criteria
for grading:
 | Includes clear hypothesis or research question (see glossary for
definitions of "hypothesis" and "research question"),
and persuasive rationale for investigation |
 | Articles listed are reports of original research (see glossary for
definitions of "research"; use Peritz or Shera), and are relevant
to stated problem |
 | Annotations are clear, informative |
Needless to say, your paper should have your name and a
good title at the top of the first page, and if you submit it electronically,
you should give it a file name that includes your name as well as the
assignment. "ChrisMarloweBib557" would be much better than just
"Bib557" or "Bibliography."
Rubric to be used in evaluating and grading your work
Introduction: 30%
 |
Exceeds the standard (B+ and up): the topic is useful and well-framed.
The problem statement develops a convincing rationale for studying the
topic, weaving relevant research and non-research literature into a
cohesive account of the current state of affairs. The purpose statement
states what variables are to be investigated, defines them in
operational terms, and specifies any relationships the writer expects to find between them. The writer demonstrates a strong understanding of professional
relevance and research logic. |
 |
Meets the standard (B): the topic is relevant and clear. The problem
statement places the topic adequately in either a professional or a
research context, or both. The purpose statement is clear, explaining
what research questions or hypotheses are to be investigated. |
 |
Approaches the standard (B- and below): the topic may be irrelevant
or minimally relevant to
the course. The problem statement may be sketchy
and unconvincing, or may concentrate on the author's personal needs and
tastes rather than on the needs and tastes of the library public. The
purpose statement may be unclear, and may demonstrate misunderstandings
of the logic of research (e.g., mistaking the nature of variables and
how they could be related to each other). |
Selection: research studies evaluated:
30%
One snag here: identifying what is actually research. What is the difference
between a case study and a "how we do it real good at our library"
article? Remember these two tests:
- Purpose: research is undertaken in order to add to or test
human knowledge; if the activities reported were undertaken for other reasons and then
just seemed worth reporting, it's probably not research.
- Method: research is systematic; if there's not a clear explanation of how data were
collected and analyzed, it's probably not research. When in doubt,
ask.
There should be a minimum of 5 related research studies;
non-research articles, which may be cited as relevant in the introduction or
conclusion, do not count toward this total. Studies evaluated should be taken whenever possible from peer reviewed
research journals (like Library and Information Science Research, J of
the American Society for Information Science, or College and Research
Libraries), and not from trade magazines (like American Libraries
or Library Journal) that may popularize the results of research
without describing its methods. Studies evaluated should contribute to resolving the question or testing
the hypothesis formulated in the purpose statement.
 |
Exceeds the standard (B+ and up): more than 5 studies; studies are fully relevant to the research questions
or hypotheses formulated in the purpose statement; studies are
significant, and are taken from peer reviewed journals |
 |
Meets the standard (B): 5 studies, relevant to the purpose statement,
and taken from peer reviewed journals |
 |
Approaches the standard (B- and below): fewer than 5 studies; studies
not actually research; studies drawn from readily accessible but less
authoritative journals or online sources; studies not germane to
research question or hypothesis |
Annotations: 30%
For the examples below, assume I'm doing a literature review on the general
question of "How can GSLIS better meet the needs of students in on-line
courses?" Obviously, that question needs refining; I'd need a profile of
the students and an operational definition for their needs, for starters. I
might start my literature review with too broad a question, and define it better
as I go. And even if I put the article by Collins and Veal (2004) in my
annotated bibliography, I might decide it didn't belong in my paper.
Each annotation should describe the study's methods of gathering
and analyzing data, as well as its findings. What you want is not
necessarily a long description, but enough information so that your reader can
understand the basis for your evaluation. Example: "Collins and Veal (2004)
had a convenience sample of 143 off-campus adult learners complete two
questionnaires: Bostick's Library Anxiety Scale, and an Attitude Toward
Educational Use of the Internet scale developed by Duggan et al. Using
regression analysis, Collins and Veal determined that two factors associated
with library anxiety -- knowledge of the library and affective barriers --
accounted for 9% of the variance in respondents' attitude toward the
Internet."
Each annotation should evaluate the study largely on the basis of its methods.
Example: "Although the statistical sophistication of this study is
impressive and the questionnaires had been validated in previous studies, the
sample of adult learners was small and non-random. We should be cautious in
generalizing this estimate of the relationship between library anxiety and
attitudes toward Internet research."
Annotations should also evaluate studies for their contribution to answering
the student's research questions. Example: "If we replicated the
Collins and Veal study with MLIS students as participants, the effect size might
be even smaller than 9%; few MLIS students are likely to have serious cases of
library anxiety, but observation shows a much wider variation in their attitudes
toward the Internet. This study gives us limited guidance in helping our adult
MLIS students use their library skills to overcome Internet-related anxiety and
skills deficits."
 |
Exceeds the standard (B+ and up): studies are described in sufficient
but not excessive detail; both findings and methods are clearly
understandable in summary. Studies are evaluated both in their own right
and in relation to the student's research question, and it is easy to
understand which studies support any given response to the research
question. Evaluation of generalizability or transferability is well
grounded in a logical understanding of research methods. |
 |
Meets the standard (B): findings and methods of studies are adequately
summarized; it is clear what research method or methods were used in
each study; studies are intelligently related to the research question. |
 |
Approaches the standard (B- and below): studies may be described
incompletely (e.g., findings may dominate, and important aspects of the
research method, like survey response rate, may be omitted). It may be
impossible to tell from the annotation whether or not the article was
really reporting a research study. Studies may
be evaluated on subjective grounds, without regards to methodological
adequacy (e.g., "this is a useful study because it shows us how
important good service is"; "this study is not valuable
because it is too hard to read and too theoretical"). Little
attempt may be made to evaluate generalizability or transferability, and
student's understanding of research methods may seem shaky. (If your
understanding IS shaky, ask for help early and often. That's what you're
supposed to be learning in this course!) The contribution of each study
to answering the paper's central question(s) may be unclear. |
Mechanics: 10%
 |
Exceeds the standard (B+ and up): almost no
typos or grammatical errors. Research terms and statistics are used
appropriately. Organization is clear and easy to follow; annotations are
given in a logical order (e.g., alphabetical, chronological, or grouped
by methods or findings) and the order is explicitly stated. Citations are correct,
complete, consistent, and made whenever appropriate -- not just when
there's a direct quotation. The paper is handed in on time, unless the
student experiences an unpredictable emergency that prevents it (such as
an illness, or a family tragedy). |
 |
Meets the standard (B): few typos or
grammatical errors, and any that occur do not interfere with
comprehensibility. Citations are correct, complete, and made whenever
appropriate -- not just when there's a direct quotation. The paper is
handed in on time, unless the student experiences an unpredictable
emergency that prevents it (such as an illness, or a family tragedy), or
makes suitable arrangements in advance for predictable problems (such as
a trip to Bermuda or an onslaught of perfectionism resulting in writer's
block). |
 |
Approaches the standard (B- and below):
paper may be poorly proofread; typos
and poor grammar may require translation. Citations may be incorrect
(e.g., with authors' names spelled wrong), incomplete (e.g., with volume
or page numbers missing), inconsistent (e.g., some with the date between
the author and the title, others with the date after the publisher), or
simply not given at all when called for; the annotations may not be in
any clear order. The paper may be late with no
acceptable excuse. |
|