| HOME | HELP | FEEDBACK | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
||||||
|
||||||||
ESSAY |
* The Department of Information Science, Bar-Ilan
University, Ramat-Gan 52900, Israel; The
Bioinformatics and Biological Computing Unit, Department of Biological
Services, The Weizmann Institute of Science, Rehovot, Israel;
The Bauer Center for Genomics Research,
Harvard University, Cambridge, MA 02138
Submitted November 29, 2004; Revised February 9, 2005; Accepted March 14, 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
Key Words: instructional design theory • Gagne • conditions of learning • bioinformatics • graduate
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
The need to support bioinformatics education has been widely recognized by scientists and industry, as well as by government institutes (e.g., Altman, 1998; Brass, 2000; Gavaghan, 2000; MacLean and Miles, 1999). In a 1998 report submitted to the White House Office of Science and Technology Policy, it was declared that "There is a national need for training and education in bioinformatics" (Bioinformatics in the 21st century, 1998). In 2001, the National Institutes of Health (NIH) and the National Science Foundation (NSF) conducted a workshop in an attempt to assess needs in bioinformatics research, training, education, and career development and to develop a list of recommendations to address identified gaps (Swaja et al., 2001, p. 1). In Israel, where the program described in this essay took place, the Ministry of Science and Technology supports a national Center of Knowledge for Bioinformatics Infrastructure (COBI), which provides training, consultation, and support services and maintains infrastructure for bioinformatics research (Center of Knowledge for Bioinformatics Infra-structure, 2004). Editorials (e.g., Brass, 2000; Gavaghan, 2000; Pearson, 2001) and scientific conferences (e.g., Workshop on Education in Bioinformatics) also discuss bioinformatics education.
The literature on bioinformatics education often covers topics at a macro level, such as integrating bioinformatics into undergraduate and graduate programs, the desired contents of bioinformatics curricula (Altman, 1998; Feig and Jabri, 2002; Honts, 2003; Salter, 1998), what audiences should be trained, and what resources should be devoted to bioinformatics education (Swaja et al., 2001). Others provide examples of courses and ongoing bioinformatics programs (e.g., Altman and Koza, 1996; Campbell, 2003; Feig and Jabri, 2002; Jenkins, 2000; Kim, 2000; Magee et al., 2001). Delivery methods, especially distance learning, are also a major concern (e.g., Brass, 2002; Cheng, 2002).
Developing new and better instructional methods is one of the challenges facing bioinformatics educators and support services (Ben-Dor et al., 2003). Yet this issue has been largely overlooked in the scholarly discourse on bioinformatics education. A few exceptions include Abbot (2002), Cheng (2002), Choo et al. (2004), Courtois and Handel (1998), and Kim (2000). These exceptions stress the dearth in the literature rather than satisfy the need to develop better instructional methods for bioinformatics.
In this paper we report a systematic attempt to design bioinformatics training on the basis of Robert Gagne's Conditions of Learning instructional design theory (Gagne, 1977; Gagne and Briggs, 1974). To better assess the feasibility of applying this theory to bioinformatics training, two workshops were designed: a microarray analysis workshop (Shachak et al., 2003) and a primer design workshop. We begin this paper by reviewing the characteristics of instructional design theories in general and their plausible importance for bioinformatics education. Then a brief description of Gagne's theory is provided. The instructional design process is illustrated using examples from the two workshops. Finally, some qualitative empirical findings are provided and the applicability of Gagne's theory to bioinformatics education is discussed.
|
BACKGROUND |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
As the definition above implies, applying instructional design theories might improve teaching and learning in general. This is especially important for bioinformatics, since bioinformatics poses special challenges to instructors and developers of educational and training programs. First, unlike many other topics, bioinformatics education requires hands-on exercises. Sitting in a classroom and listening to lectures is certainly not sufficient to qualify biologists in the use of bioinformatics applications (Ben-Dor et al., 2003). Second, bioinformatics combines exact sciences (mathematics, computer science) and an empirical science (biology). Therefore, it requires biologists not only to adopt the use of new laboratory and information tools, but also to integrate theories and models rooted in other disciplines. The use of instructional design theories might help bioinformatics instructors to face these challenges by structuring educational activities and adapting teaching methods to various situations and cognitive learning processes.
Gagne's Conditions of Learning Instructional Design Theory
Conditions of Learning is an instructional design theory developed by
Robert M. Gagne as an attempt to make teaching a systematic, rule-guided
process (Zemke, 1999).
Although first published more than 35 years ago, it is still fundamental in
the fields of instructional design and instructional technology
(Merrill, 2000); parts of it
were included in newer theories, such as Instructional Transaction Theory
(Merrill, 1999).
Gagne (1977, p. 3) defines learning as "a change in human disposition or capability which persists over a period of time, and which is not simply ascribable to process of growth." Taking a behaviorist perspective, he argues that the best indication that learning has occurred is a change in performance. Performances can be categorized into five domains of learned capabilities: motor skills, intellectual skills, cognitive strategies, verbal information, and attitudes. According to the Conditions of Learning theory, different internal and external conditions are necessary for each learned capability. For example, to learn a new motor skill the learner must acquire other motor skills of which it is composed (internal condition) and receive feedback (external condition). Lower-level intellectual skills (e.g., concepts, rules) and information must be acquired to learn high-level intellectual skills (e.g., problem solving). Regardless of the domain of performance, Gagne (1977) argues there are eight basic forms of learning that compose all learning performances. Those basic forms of learning are extracted from previous works of behaviorists such as Pavlov, Skinner, and Thorndike (for review, Zemke, 1999).
The most significant contribution of Gagne's theory, however, is nine instructional events outlined by it. Each corresponds to a different cognitive process (Zemke, 1999). These instructional events should satisfy or provide the necessary conditions for learning and serve as the basis for designing instruction and selecting appropriate media (Gagne, 1977). The nine events are: 1) gaining attention, 2) informing the learner of the objective, 3) stimulating recall of prior knowledge, 4) presenting the stimulus, 5) providing learning guidance, 6) eliciting the performance, 7) providing feedback about performance correctness, 8) assessing performance, and 9) enhancing retention and transfer (Gagne and Briggs, 1974).
These three components of Gagne's theory—learned capabilities, the basic forms of learning, and instructional events—provide the basis for defining learning objectives and creating a sequence of instruction (Gagne and Briggs, 1974). It will be further discussed and illustrated in the following sections.
|
WORKSHOPS DESIGN PROCESS |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
|
Step 1: Defining Learning Objectives
The first step in developing training materials on the basis of Gagne's
Conditions of Learning was to define top-level learning objectives for each
workshop. Learning objectives describe what the learner should be able to do
after completing the instruction. Thus, they define the type of learned
capability or, in other words, the domain of performance.
According to Gagne, the only proof of learning is a change in performance. In order to clearly define the desired change in performance, learning objectives have to be observable and measurable. Gagne and Briggs (1974) suggest that all learning objectives include the following five elements: situation, learned capability, object, action, tools and other constraints. Though this exact scheme was not followed, the defined learning objectives contained most of these elements. Example 1 from the microarray analysis workshop demonstrates that.
Example 1: Learning Objective of the Microarray Analysis Workshop.
[situation]Given gene expression data, [object] the learners [learned capability: problem solving] will be able to correctly analyze the data [tools and constraints] using GeneSpringeTM software.
In this example, the learned capability is described in general terms "correctly analyze." To make this objective observable and measurable, evaluation criteria for correct analysis must be defined. Specifying learning objectives for prerequisite information and skills, as described below, can help in defining such criteria.
Step 2: Creating a Hierarchy of Objectives
The design of learning objectives is a top-to-bottom process. Example 1
above demonstrates a learned capability of higher-level intellectual skill
(i.e., problem solving). Learning of intellectual skills requires that
"presentation of learning situation for each new skill should be
preceded by prior mastery of subordinate skills"
(Gagne and Briggs, 1974, p.
105) and that "information relevant to the learning of each new skill
should be previously learned or presented in instruction"
(Gagne and Briggs, 1974, p.
105). Therefore, the next step was to define learning objectives for
prerequisite information and skills. For each of these prerequisite
objectives, the process repeated itself until reaching a level assumed to be
known by all students.
The following example from the microarray analysis workshop demonstrates this principle: in order to be able to perform microarray data analysis, one needs to measure and reduce the noise contained in microarray experimental data. This implies a learned capability of applying rules. To do that, one needs to know what creates the noise in microarray data and discriminate systemic noise from random noise. This defines learned capabilities of acquiring concepts (systemic/random noise) and recall of information. The same principle applies for methods, data, and biological questions. In addition, all four "branches" are interconnected. The result of this process is a complex hierarchy,1 as shown in Figure 2. This hierarchy of objectives provides a general framework for the microarray analysis workshop instructional sequence.
|
Hierarchies of objectives were created for the two workshops. As described below, these hierarchies were used to design lectures and modify existing hands-on tutorials—a previously designed tutorial for the microarray analysis workshop (Ron Ophir, unpublished), and the commercial tutorial provided with the Oligo®6 demo software for the primer design workshop.
Step 3: Incorporating the Events of Learning into the Hands-on Tutorials
At this step, events of learning proposed by the Conditions of Learning
theory were incorporated into the modified hands-on tutorials.
Figure 3 demonstrates the
incorporation of some learning events into the primer design hands-on
tutorial.
|
The following three events were deliberately incorporated into the tutorial manual: 1) informing the learner of the objective, 2) stimulating recall of prerequisite learning, and 3) providing feedback about performance correctness. The last two of Gagne's proposed events—assessing the performance and enhancing retention and transfer—were ignored because they did not fit with the voluntary nature of participating in the workshops or with the limited time frame. Incorporation of the three learning events mentioned above resulted in a first draft of the hands-on tutorial, which was tested and revised as described below.
Step 4: Testing and Revising the Tutorials
At this stage, the first draft of the tutorials was tested on a small
number of trainees and revised. The microarray analysis tutorial was tested on
a group of students participating in a microarray laboratory course. Two
groups of two and three students were closely observed. For the primer design
tutorial, it was impossible to perform the test in a real classroom situation.
Therefore, it was tested by two volunteers. For each tutorial, the following
information was recorded: 1) inconsistencies between the tutorial and the
screen, 2) questions addressed to the instructor, 3) comments on the work
process made by the students to each other and to the
observer,2 4) errors made by the
learners,3 and 5) suggestions for
changes in the tutorial.
These observations allowed us to find errors and improve the tutorials. For example:
1. Some of the orders in the microarray analysis tutorial were phrased "write x,y,z in the file name line." One observant commented that it should be reversed: "in the file name line write x,y,z." This comment reflected that the natural cognitive process is first to identify the window elements, then to move the cursor to the right position, and finally to write the file name. Therefore, the tutorial was revised to support this cognitive process.
2. One of the volunteers who tested the primer design tutorial commented that working with the tutorial made him feel he did not complete the tasks. Thus, the sequence of tasks was modified to reflect a complete and systematic process.
Step 5: Conducting the Workshops
The final tutorials were employed in three 1-day workshops: one microarray
analysis workshop conducted at the Weizmann Institute of Science, Israel, and
two primer design workshops conducted at the Weizmann Institute and Tel-Aviv
University, Israel. Each workshop consisted of introductory lectures and a
hands-on session conducted in a computer laboratory. About 80 people
participated in the introductory lectures for the microarray analysis
workshop. Of them, 18 participated in the hands-on session. A total of 36
people attended the primer design workshops: 15 at the Weizmann Institute and
21 at Tel-Aviv University. All of them also took the hands-on session. During
the workshops participants were asked to fill in questionnaires that included
data on demographic variables (gender, age, degree, and role) and
self-reported level of experience in using computers and bioinformatics
applications. Fifteen participants in the microarray analysis workshop and 30
in the primer design workshops completed the questionnaires (response rates of
83.3 percent and 66.7 percent, respectively). Descriptive statistics for the
two groups are presented in Table
1.
|
During the hands-on session of the microarray analysis workshop, 5-min observations of several groups were conducted. The main findings of these observations were:
In the primer design workshops it was impossible to conduct systematic observations. However, in one of these workshops the instructor's impression was similar to the above findings. In the other workshop, the impression was that most students did follow the tutorial step-by-step, without skipping any part of it. In that workshop, students rarely made errors and seldom needed the instructor's assistance. The implications of these findings are discussed below.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
As mentioned earlier, newer instructional design theories build on Gagne, or have adopted some components of his theory. However, Conditions of Learning and related theories also have been criticized for being "very inefficient to use because an instructional designer must build every presentation from fundamental components" (Merrill et al., 1991). Our personal experience with designing two different bioinformatics tutorials does not support this criticism. From a designer's perspective we conclude that Conditions of Learning provides a practical framework for developing bioinformatics training in an academic setting. Specifically, defining learning objectives, top-to-bottom design, and creating a hierarchy of objectives were useful in defining priorities, structuring training activities, dividing the tutorial into units, and determining the sequence of instruction.
As a way of teaching, however, there are mixed findings regarding the effectiveness of the theory. While students in one of the primer design workshops followed the tutorial step-by-step, did not skip any parts of it, and hardly made errors, students in the two other workshops often skipped long reading paragraphs, tried to learn by exploration, and often needed the instructor's assistance to recover from errors. The instructors, especially in the microarray analysis workshop, felt they had spent more time answering questions, helping people recover from errors, and explaining issues already described in the tutorials than they initially expected.
A possible explanation for the observations in the microarray analysis workshop and one of the primer design workshops is that participants were highly experienced learners. As seen in Table 1, all participants had a B.S. degree or higher, they had more than average computer experience, and the vast majority of them had working experience with at least one bioinformatics tool, most commonly BLAST—87 percent of the microarray analysis and 83 percent of the primer design trainees used it. It is possible that such experienced learners prefer learning by exploration over structured step-by-step, although this way of learning is not supported by the tutorials described here. A similar proposition was made by researchers in human-computer interaction, who studied the way office workers learned word processing (Carroll et al., 1988). Nevertheless, the difference between the various workshops remains unexplained.
Instructional design theorists proposed that instruction based on Conditions of Learning and on related theories is often passive rather than interactive (Merrill et al., 1991). As discussed above, bioinformatics education requires interactivity in using computer applications. Together with the observations and instructors' perceptions described above, this criticism suggests that Conditions of Learning is not an optimal way to teach bioinformatics. Other learning and instructional design theories, which better support exploration and interactivity, may be sought. Some promising directions are Problem Based Learning (for review, Allen and Tanner, 2003), which has been recently applied to bioinformatics education (Choo et al., 2004), and Minimalism—a theory derived from the field of human-computer interaction, which was specifically designed to support the way people learn to use computer applications (Carroll et al., 1987-8). The feasibility and effectiveness of applying these approaches to bioinformatics teaching has yet to be verified.
|
LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
Another possible direction for future research is that proposed at the end of the Discussion section: to study the effectiveness of exploration-based approaches for teaching bioinformatics such as Problem Based Learning (Allen and Tanner, 2003) or Minimalism (Carroll et al, 1987-8).
|
ACCESSING MATERIALS |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 The term hierarchy used by Gagne and Briggs
(1974) does not necessarily
imply a linear sequence of instruction. Rather, it often refers to the
organization of learning objectives in a complex, networked structure as is
the case here. 
2 The students had been informed of the observation's purpose and knew they
were being watched. Though they were not requested to, the students often made
comments and suggestions to the observer about the tutorial.
3 Errors may be categorized as semantic errors, which arise from the
use of inappropriate method in a particular situation; syntactic
errors, which occur when a correct action is used improperly; and
slips, such as typing mistakes
(Lazonder and van der Meij,
1995). For example, in GeneSpringeTM, folders are displayed
on the main screen navigation panel, but also in dialog windows that open. A
common semantic error was selecting a folder from the main navigation panel
instead of from the dialog window. In Oligo®6, a common syntactic error
was that after defining the ranges in which primers are searched, people
clicked "OK" instead of defining search parameters by clicking
"Parameters." As a result, the display on the screen was different
than that described in the tutorial.
Address correspondence to: Aviv Shachak (shachaa4{at}popeye.os.biu.ac.il).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDWORKSHOPS DESIGN PROCESSDISCUSSIONLIMITATIONS AND DIRECTIONS FOR...ACCESSING MATERIALSREFERENCES |
|---|
Allen, D., and Tanner, K. (2003). Approaches to cell biology teaching: learning content in context—problem-based learning.Cell Biol. Educ. 2(2),73 -81.[Medline]
Altman, R.B. (1998). A curriculum for bioinformatics:
the time is ripe. Bioinformatics
14(7),549
-550.
Altman, R.B., and Koza, J. (1996). A programming course in bioinformatics for computer and information science students. InProc. Pac. Symp. Biocomput.73 -84.
Ben-Dor, S., Butcher, S., Iyer, K.L., Kelso, J., Littlejohn, T., Shachak, A., and Rubin, E. (2003). Training and support for bioinformatics: theoretical and practical aspects. Paper presented at theIntelligent Systems for Molecular Biology (ISMB) conference , Brisbane, Australia.
Bioinformatics in the 21st century. (1998). A report to the Research Resources and Infrastructure Working Group, Subcommittee on Biotechnology, National Science and Technology Council, White House Office of Science and Technology Policy. http://clinton4.nara.gov/WH/EOP/OSTP/NSTC/html/bioinformaticsreport.html#Training (accessed 26 October 2004).
Brass, A. (2000). Bioinformatics education—a UK
perspective. Bioinformatics
16(2),77
-78.
Brass, A. (2002). Experiences from developing and delivering the current programme in bioinformatics by distance learning. Presentation at the Workshop on Education in Bioinformatics (WEB), Edmonton, Canada.
Campbell, M.A. (2003). Public access for teaching genomics, proteomics, and bioinformatics. Cell Biol. Educ. 2(2),98 -111.[Medline]
Carroll, J.M., Mack, R.L., Lewis, C.H., Grischkowsky, N.L., and Robertson, S.R. (1988). Exploring a wordprocessor. In:Effective Documentation: What We Have Learned from Research , S. Doheny-Farina, ed. Cambridge, MA: The MIT Press,103 -126.
Carroll, J.M., Smith-Kerker, P.L., Ford, J.R., and Mazur-Rimetz, S.A. (1987-8). The minimal manual. Hum. Comput. Interact. 3,123 -153.[CrossRef]
Center of Knowledge for Bioinformatics Infrastructure (2004). http://cobi.org.il/ (accessed 26 October 2004).
Cheng, B. (2002). The third degree: distance education and bio-medical informatics. Presentation at the Workshop on Education in Bioinformatics (WEB), Edmonton, Canada.
Choo, K.H., Tong, J.C., Tan, T.W., and Ranganathan, S. (2004). Application of Problem Based Learning pedagogy in global online bioinformatics education Georgia Tech. Presentation at theWorkshop on Education in Bioinformatics (WEB) , Glasgow, Scotland.
Courtois, M.P., and Handel, M.A. (1998). A collaborative approach to teaching genetics information sources. Res. Strat. 16(3),211 -220.
Feig, A.L., and Jabri, E. (2002). Incorporation of bioinformatics exercises into the undergraduate biochemistry curriculum.Biochem. Mol. Biol. Educ. 30(4),224 -231.
Gagne, R.M. (1977). Conditions of Learning, 3rd ed. New York: Holt, Rinehart and Winston.
Gagne, R.M., and Briggs, L.J. (1974).Principles of Instructional Design . New York: Holt, Rinehart and Winston.
Gavaghan, H. (2000). Europe seeks solution to bioinformatics shortfall. Nature 404(6778),687 -688.[CrossRef][Medline]
Hemminger, B. (2004). Bioinformatics program summary. http://ils.unc.edu/bmh/bioinfo/ASIST02_bioinformatics_programs_summary.html (accessed 17 November 2004).
Honts, J.E. (2003). Evolving strategies for the incorporation of bioinformatics within the undergraduate cell biology curriculum. Cell Biol. Educ. 2(4),233 -247.[CrossRef][Medline]
Jenkins, R.O. (2000). Editorial: meeting the demand for bioinformaticians. Biochem. Mol. Biol. Educ. 28,272 -273.[CrossRef]
Kim, T.D. (2000). PCR primer design: an inquiry-based introduction to bioinformatics on the World Wide Web. Biochem. Mol. Biol. Educ. 28,274 -276.[CrossRef]
Lazonder, A.W., and van der Meij, H. (1995). Error-information in tutorial documentation: supporting users' errors to facilitate initial skill learning. Int. J. Hum-Comput. St. 42(2),185 -206.[CrossRef]
Luo, J. (2002). Bioinformatics service, education and research: the EMBnet and CBI. In Silico Biol. 2(0016). http://www.bioinfo.de/isb/2002/02/0016/main.html (accessed 26 October 2004).
MacLean, M., and Miles, C. (1999). Swift action needed to close the skills gap in bioinformatics. Nature 401(6748),10 .[Medline]
Magee, J., Gordon, J.I., and Whelan, A. (2001). Bringing the human genome and the revolution in bioinformatics to the medical school classroom: a case report from Washington University School of Medicine.Acad. Med. 76(8),852 -855.[Medline]
Merrill, M.D. (1999). Instructional Transaction Theory (ITT): instructional design based on knowledge objects. In:Instructional-Design Theories and Models: A New Paradigm of Instructional Theory , Vol. 2, ed. C.M. Reigeluth. Mahwah, NJ: Lawrence Erlbaum Associates,397 -424.
Merrill, M.D. (2000). Suggested self-study program for instructional systems development (ISD). http://www.id2.usu.edu/MDavidMerrill/IDREAD.PDF (accessed 17 November 2004).
Merrill, M.D., Li, Z., and Jones, M.K. (1991). Limitations of first generation instructional design (ID1). Educ. Technol. 30(1),7 -11.
Pearson, W.R. (2001). Editorial: training for
bioinformatics and computational biology. Bioinformatics
17(9),761
-762.
Reigeluth, C.M. (1999). What is instructional-design theory and how is it changing? In: Instructional-Design Theories and Models: A New Paradigm of Instructional Theory, Vol.2 , ed. C.M. Reigeluth. Mahwah, NJ: Lawrence Erlbaum Associates, 5-29.
Salter, H. (1998). Teaching bioinformatics.Biochem. Educ. 26(1),3 -10.
Samish, I. (2003). Bioinformatics education in Israel: academic, private and unique retraining programs. Presentation at theWorkshop on Education in Bioinformatics (WEB03) , Brisbane, Australia.
Shachak, A., Ophir, R., Rubin, E., and Fine, S.F. (2003). Instructional-design theories in bioinformatics education: an application of Gagne's "Conditions of Learning" to microarray analysis training. Presentation at the Workshop on Education in Bioinformatics (WEB), Brisbane, Australia.
Swaja, R.E., Rastegar, S., Griffith, L.G., Sachs, M.B., and Subramaniam, S. (2001). Assessing bioengineering and bioinformatics research training, education and career development: opportunities for NIH and NSF collaboration. http://www.bisti.nih.gov/NSFNIHFinalReport824.pdf (accessed 17 November 2004).
Zemke, R. (1999). Toward a science of training.Training 36(7),32 -36.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | ARCHIVE | SEARCH | TABLE OF CONTENTS |
