Biology Coursework Skill Deficit

Abstract

Regrettably, the sciences are not untouched by the plagiarism affliction that threatens the integrity of budding professionals in classrooms around the world. My research, however, suggests that plagiarism training can improve students' recognition of plagiarism. I found that 148 undergraduate ecology students successfully identified plagiarized or unplagiarized paragraphs three-quarters of the time. The students' ability to identify plagiarism was not significantly different when the quoted or paraphrased text included complex sentence structure and scientific jargon and when it included only simple sentences that mostly lacked jargon. The students who received plagiarism training performed significantly better at plagiarism detection than did those who did not receive the training. Most of the students, independent of training, identified properly paraphrased, quoted, and attributed material but had much greater difficulty identifying paraphrases that included long strings of copied text—up to 15 words—or proper paraphrases that lacked citations. The misunderstanding of paraphrasing and citation conventions found here could manifest as unintentional plagiarism in these students' later work.

Numerous self-reported rates of past cheating behaviors by students, including plagiarism, hover around 50% (Hale 1987, Franklyn-Stokes and Newstead 1995, McCabe et al. 2001). On one hand, plagiarism may be a premeditated act of deception, whereby students knowingly present the words or ideas of others' as their own (Howard 1995, Yeo 2007). Many institutions have incorporated penalties and honor codes to combat or prevent such actions; accordingly, honor codes have been shown to lower the incidence of cheating, yet they do not eliminate it (McCabe et al. 2001). Despite these measures, some students rationalize this misconduct because their perceived alternative—failure—seems worse (Park 2003, Power 2009). On the other hand, some students are unaware that their actions, or absence of actions, may constitute plagiarism (Howard 1995, Park 2003).

McCabe and colleagues (2001) found that the incidence of cheating on written work was roughly identical to rates 30 years prior; however, students defined plagiarism far more loosely in 1993 than they did in 1963. For example, many students did not recognize that proper citation must accompany good paraphrasing or quoting to adequately avoid plagiarism (e.g., Wilhoit 1994, Franklyn-Stokes and Newstead 1995, Roig 1997, Rennie and Crosby 2001, Dawson MM and Overfield 2006, Yeo 2007, Power 2009). Despite the prevalence of plagiarism due to poor citation skills, this topic is often sidestepped in science curriculum. Typically, instructors warn students of plagiarism during the first week of class (Power 2009), but they provide no further instruction. In Nuss (1984), over half of the surveyed college instructors never or minimally discussed their policies on academic integrity in the classroom. Previous work (Roig 1999, Landau et al. 2002, Schuetze 2004, Belter and du Pré 2009), however, demonstrated that education can reduce unintentional plagiarism.

Student plagiarism is indisputably a cross-discipline issue. With the present study, however, I hoped to identify the source of inadvertent plagiarism in the biological sciences and to investigate how science educators can reduce it. The bulk of research on student plagiarism is derived from the social sciences (e.g., Ashworth et al. 1997, Roig 1997, Elander et al. 2010). In a notably smaller body of literature, these issues have been investigated in science, technology, engineering, and mathematics (STEM) fields, and much of that literature focuses largely on students in health fields (e.g., Julliard 1994, Gaberson 1997, Rennie and Crosby 2001). In the natural sciences, most plagiarism research is either qualitative (Willmott and Harrison 2003) or quantifies student perceptions and not their performance (Craig et al. 2010, Freeman and Lynd-Balta 2010; but see Soto et al. 2004, Dawson MM and Overfield 2006). The present study, therefore, fills a significant gap in the science education literature: It combines qualitative assessments of students' understanding of plagiarism with controlled analyses of the factors contributing to the lack of that understanding.

Specifically, my main goals were to determine how successfully biology students can identify plagiarism; whether plagiarism training improves students' ability to identify plagiarism; whether readability of the material (i.e., sentence structure, word length, and scientific terminology) affects students' ability to recognize plagiarism; and how varying the severity of plagiarism influences students' ability to discern plagiarized material from properly quoted, paraphrased, and attributed material.

Survey instrument

The course used for this study, General Ecology, is a required course for 10 majors within two colleges (the Colleges of Science and Natural Resources) at a public postsecondary institution in the western United States. The prerequisites for the course ensure that most students have previously taken General Biology for biology majors.

For this study, surveys were administered the first and last week of the fall 2010 and spring 2011 semesters through an electronic management system (Blackboard, Inc., Washington, DC). All General Ecology students during these semesters were asked to participate in the study. Survey completion constituted 1% of the students' final grade, but they could complete a short alternative assignment if they chose not to participate. The electronic survey was designed and the data were collected using Survey Monkey (www.surveymonkey.com). The full survey (see the supplemental material, available online at http://dx.doi.org/10.1525/bio.2012.62.6.9) consisted of 23 questions and was adapted from Roig (1997).

The survey gathered unique identifiers, information on each student's prior education related to plagiarism, and demographics. The remaining half of the survey was aimed at assessing the students' success or lack thereof in recognizing plagiarism. I selected two review papers related to trophic cascades (see Licht et al. 2010 and Pace et al. 1999). Secondary literature sources were chosen over primary literature because they better lent themselves to free- standing, two-sentence excerpts in which the reader, a second-year undergraduate student, would need little background knowledge to follow the content. This exercise (i.e., reading literature on theoretical concepts in ecology and summarizing them in their own words) was a key learning objective for both the 2010 and 2011 classes.

For the survey, the students were required to read the original excerpt of each paper (table 1). Then, the students were presented with six short paragraphs that were based on each excerpt (see the supplemental material). The new paragraphs represented a range of plagiarism severity (table 2), in which four paragraphs were plagiarized and two were not. The students were asked to rate each new paragraph as plagiarized or not plagiarized. The response variable—student success—was the proportion of correct responses (plagiarized when it was plagiarized or not plagiarized when it was not).

Table 1.

Original excerpts used in the survey instrument.

Difficult readability paragraph Moderate readability paragraph 
Source Pace et al. 1999Licht et al. 2010
Excerpt “Within communities, diversity and species replacement should provide a means for restricting or reducing predatory impact and hence trophic cascades. Microbial communities would appear to have significant potential to dampen cascades via rapid succession to predation-resistant forms, depending on the potential diversity of the particular group” (Pace et al. 1999, p. 486). “Although fences are not used to manage predators in natural areas in the United States, the country does have a population of wolves that functions as if it were in a large, fenced reserve: the small population of wolves on Isle Royale in Lake Superior. The 550-[kilometer] island has supported a mean of 24 wolves since 1959” (Licht et al. 2010, p. 150). 
Readability 46 words, average of 23 words per sentence, 6.4 characters per word, Flesch reading ease score = 0.0. 58 words, average of 29 words per sentence, 4.4 characters per word, Flesch reading ease score = 44.6. 
Difficult readability paragraph Moderate readability paragraph 
Source Pace et al. 1999Licht et al. 2010
Excerpt “Within communities, diversity and species replacement should provide a means for restricting or reducing predatory impact and hence trophic cascades. Microbial communities would appear to have significant potential to dampen cascades via rapid succession to predation-resistant forms, depending on the potential diversity of the particular group” (Pace et al. 1999, p. 486). “Although fences are not used to manage predators in natural areas in the United States, the country does have a population of wolves that functions as if it were in a large, fenced reserve: the small population of wolves on Isle Royale in Lake Superior. The 550-[kilometer] island has supported a mean of 24 wolves since 1959” (Licht et al. 2010, p. 150). 
Readability 46 words, average of 23 words per sentence, 6.4 characters per word, Flesch reading ease score = 0.0. 58 words, average of 29 words per sentence, 4.4 characters per word, Flesch reading ease score = 44.6. 

View Large

Table 2.

Explanation of the severity of plagiarism expressed in the new paragraphs created from each original excerpt.

Plagiarism severityaSurvey instrumentbQuotation or paraphrasecAttribution 
Good1P6, L3 Properly paraphrased Proper attribution 
Good2P2, L2 Properly quoted Proper attribution 
Fair1P1, L5 Patchwriting plagiarism Proper attribution 
Fair2P5, L4 Properly paraphrased No attribution 
Poor1P4, L1 Direct plagiarism Improper attribution 
Poor2P3, L6 Patchwriting plagiarism No attribution 
Plagiarism severityaSurvey instrumentbQuotation or paraphrasecAttribution 
Good1P6, L3 Properly paraphrased Proper attribution 
Good2P2, L2 Properly quoted Proper attribution 
Fair1P1, L5 Patchwriting plagiarism Proper attribution 
Fair2P5, L4 Properly paraphrased No attribution 
Poor1P4, L1 Direct plagiarism Improper attribution 
Poor2P3, L6 Patchwriting plagiarism No attribution 

View Large

Four factors related to the understanding of plagiarism

Plagiarism training was the primary factor of interest in this study. The fall 2010 class (n = 94) was given four writing assignments. These assignments were designed to improve the students' library literacy, to define plagiarism and allow the students to practice avoiding it, to help the students develop skills in paraphrasing, and to help the students to assemble these skills into a two-page research paper related to the ecological topics covered during class. The expected learning outcomes of these assignments were that the students would be able to identify scientific literature, to discriminate primary from secondary literature, to generate appropriate citations according to the Council of Science Editors (CSE), and to construct quotations and paraphrases free of plagiarism. The ultimate goal of this training was not to prepare the students to write primary research manuscripts or to review articles suitable for publication but, rather, to write university assignments not intended for publication. The plagiarism assignment specifically provided definitions of plagiarism and guidelines for and examples of proper quotation, proper citations (according to CSE), and proper paraphrasing (see the supplemental material). The assignment also required that the students demonstrate proficiency in plagiarism avoidance; it was adapted from a similar assignment by Paul C. Smith at Alverno College.

The spring 2011 class (n = 54) was assigned three writing assignments, all of which were different from those given to the 2010 class. Each 2011 writing assignment was a figure set designed to improve the students' math and interpretation skills through the use of ecological data (Fortier 2002, D'Avanzo and Musante 2004a, 2004b). For each assignment, the students communicated their ideas in writing, which, in most cases, required them to quote, paraphrase, or provide attribution. A formal definition of plagiarism, however, was provided on the syllabus only as a Web link to the University's policy on academic integrity. A single example of proper paraphrasing and attribution was available on the course Web site, which the students were not required to consult. Guidelines on proper quoting, citing, and paraphrasing practices were not part of the instruction or assignment materials. Several weeks into the semester, the instructor conducted a 10-minute demonstration on improper paraphrasing techniques (e.g., cutting and pasting from the Internet).

I considered the fall 2010 class to be the plagiarism- training group, because the students received the definition of plagiarism along with the guidelines and examples of how to avoid it (through quoting, paraphrasing, and citing); the students further practiced these skills in a focused assignment (see the supplemental material). The spring 2011 class was considered to be the no-plagiarism-training group, because the students received no clear definition of plagiarism and no guidelines on how to avoid plagiarism as part of the ecology class (although a few examples were provided). The students were indirectly required to practice avoiding plagiarism; that is, the goal of their assignments was to develop critical thinking and synthesis skills while also avoiding plagiarism.

To account for the inherent variability in the students' experiences with plagiarism, change over time was the second factor. The survey links were open for only the first and last weeks of the fall 2010 and spring 2011 semesters, which allowed for repeated measurements. As part of the survey, each student created a unique identifier, which protected his or her anonymity and yet allowed for longitudinal comparisons of his or her survey responses.

Previous work suggests that greater readability minimizes the incidence of plagiarism (Roig 1999, Walker 2008). As an extension, I predicted that the students would have greater success in identifying plagiarism with a more readable text. Reading ease can be evaluated using the Flesch reading ease score (Flesch 1948). This score, which ranges from 0 to 100, is derived from the average sentence length and average number of syllables per word (Flesch 1948). In the survey instrument, two levels of reading ease comprised the third factor. The first original excerpt (Pace et al. 1999) represented the difficult reading level, because it contained an average of 6.4 characters per word and had a Flesch reading ease score of 0.0 (table 1). New paragraphs derived from this excerpt, on average, had 6.5 characters per word and a Flesch score of 2.0. In addition, this excerpt contained numerous scientific terms (e.g., species replacement, predatory impact, trophic cascades) that may lie outside the average second-year biology student's vernacular. The second original excerpt (Licht et al. 2010) was considered the moderate reading level, with a Flesch reading ease score of 44.6, an average of 4.4 characters per word, and few scientific terms (table 1). New paragraphs based on this excerpt were also more readable (an average of 4.5 characters per word and a Flesch score of 50.2).

The fourth and final factor that I evaluated was the severity of plagiarism. Plagiarism may be viewed as a sliding scale, on which properly quoted, paraphrased, and attributed materials represent one end of the scale and direct quotes lacking quotation marks or paraphrases including strings of copied words accompanied by a lack of attribution represents the opposite end (table 2). I hypothesized that the students would succeed in identifying the ends of this severity scale and would be more likely to fail in identifying plagiarism in the middle (i.e., proper paraphrases with no attribution or paraphrases with strings of copied words with proper attribution). For each of the two original excerpts, six new paragraphs represent this range of plagiarism severity (table 2).

I used an analysis of variance of a four-way factorial in a split—split plot design to test the effects of the four factors and the interactions on the students' ability to accurately identify plagiarism or the lack thereof. The students were used as whole-plot units, plagiarism training as the whole-plot factor (treatment or control), time as the subplot factor (the first or last week of the semester), and question readability (moderate or difficult) and plagiarism severity (see table 2) as sub-subplot factors. I coded student success (i.e., their identification of plagiarism as plagiarized and unplagiarized work as not plagiarized) as a binary response variable and therefore used a binomial distribution in my analyses. The data analysis was generated using the GLIMMIX procedure in SAS (Version 9.22, SAS Institute, Cary, North Carolina).

Preliminary student knowledge and background

Over two semesters, 148 students participated in this study. Eighty-four percent of the participants (124 students) completed both the pretreatment and the posttreatment surveys. The majority of these students were juniors (43%) and sophomores (37%), whereas freshmen (5%) and seniors (15%) represented a small portion of the population of sampled students. Of the students that reported demographic information, 61% were female, and 39% were male. Nearly all of the students identified English as their first language (99%). Ninety percent of the participants identified their major within the Colleges of Science or Natural Resources. The self-reported average grade-point average was 3.3 on a 0.0–4.0 scale. The first week of class, during the pretreatment survey, over half of the students reported that their understanding of what constitutes plagiarism was good (53%), whereas a third indicated that their understanding was very good (34%); only 12% reported a fair understanding, and a single student reported a poor understanding. At the start of both classes, 87% of the students reported having learned about plagiarism through formal assignments in their prior education, and the remaining 13% had previously been exposed to at least definitions or examples of plagiarism.

The first week, no significant differences in student success existed between the students that would receive plagiarism training and those that would not (F(1, 122) = 0.02, p = .88). The students matriculated in General Ecology during the fall 2010 and spring 2011 semesters failed to identify plagiarism or misidentified properly quoted, paraphrased, or attributed material as plagiarism roughly a quarter of the time (for the training group, the mean (M) = 74.8%, standard error (SE) = 2.61; for the no-training group, M = 74.1%, SE = 3.96; see figure 1). This pattern was due to nearly all of the students' missing a few questions, rather than to a quarter of the students' missing all the questions (figure 2). Most of these errors occurred when the students overlooked plagiarized paragraphs, because the success rates for properly quoted, paraphrased, and attributed paragraphs were far higher (M = 88.4%, SE = 0.10). Furthermore, student success did not correlate with self-assessments of plagiarism understanding (r2 = .02).

Figure 1.

Comparison of student success rates (the proportion of correct assignment of plagiarized or nonplagiarized paragraphs) between the two survey periods and between the two classes. The fall 2010 class, which received plagiarism training (depicted with black triangles) exhibited a substantial increase in student success over time relative to the spring 2011 class, which did not receive plagiarism training (shown as gray squares). The data are backtransformed means. The error bars represent 80% confidence limits.

Figure 1.

Comparison of student success rates (the proportion of correct assignment of plagiarized or nonplagiarized paragraphs) between the two survey periods and between the two classes. The fall 2010 class, which received plagiarism training (depicted with black triangles) exhibited a substantial increase in student success over time relative to the spring 2011 class, which did not receive plagiarism training (shown as gray squares). The data are backtransformed means. The error bars represent 80% confidence limits.

Figure 2.

Cumulative frequency distribution (the proportion of values below the indicated score) of correct responses on the pretreatment survey for all of the students (the plagiarism-training group and the no-training group). The distribution suggests that all students missed a few questions; a more positively skewed distribution, which was not found in these data, would have suggested that only a quarter of the students missed all of the questions, whereas the remaining 75% missed none.

Figure 2.

Cumulative frequency distribution (the proportion of values below the indicated score) of correct responses on the pretreatment survey for all of the students (the plagiarism-training group and the no-training group). The distribution suggests that all students missed a few questions; a more positively skewed distribution, which was not found in these data, would have suggested that only a quarter of the students missed all of the questions, whereas the remaining 75% missed none.

Factors contributing to student success

Plagiarism training clearly contributed to positive student outcomes (training × time interaction, F(1, 2684) = 13.79, p < .001; figure 1). Improvement due to training was not affected by readability (training × time × readability interaction, F(1, 2684) = 3.32, p = .0686), by severity (training × time × severity interaction, F(5, 2684) = 1.73, p = .1254), or by the two in combination (training × time × readability × severity interaction, F(5, 2684) = 1.39, p = .2260). Student success also varied with plagiarism severity (F(1, 2684) = 68.93, p < .001), regardless of training or time, but was uniformly high for the good severity rating (i.e., correctly quoted, paraphrased, and attributed paragraphs).

Student-driven, inquiry-based science laboratory curricula have received increasing support as an instructional model in order to increase student interest and engagement with course material as well as to increase literacy in the nature of the scientific process and sense of responsibility for the success of their learning (Luckie et al., 2004; Frantz et al., 2006; Weaver et al., 2008). In this model students play an integral role in directing the activities in the lab by asking questions and designing experiments. Often, although the students direct the design of experiments, they are revisiting well-studied scientific questions and typically know that they are doing so (Weaver et al., 2008). Compounding this is the trove of scientific information of variable depth and reliability that is readily available through online sources. Students should evaluate this information before using it, but they may not have the skills to do so. One of the skills needed is a good understanding of how that information has come to be through the process of scientific experimentation. Without this experience in the design and execution of experiments, the ability of students to critically evaluate information is limited.

A variation on the inquiry-based approach involves students participating in novel research projects as part of their undergraduate laboratory courses (NAS Bio 2010), an approach that has received bolstered support with the 2011 AAAS/NSF Vision and Change in Undergraduate Biology Education document. Thus far, these have ranged from brief activities within a general lab course (Birkett, 2009) to partial semester research-based laboratory classes (Weaver et. al., 2006; Weaver et al., 2008) to full- semester courses (Luckie et al., 2004; Frantz et al., 2006; current study). These courses follow the steps for incorporating inquiry put forth by Concannon and Brown (2008). The students are provided with the background knowledge and skills required to formulate questions, design a novel experiment or set of experiments, carry out the experiments under the supervision of the instructional staff for the course, and analyze and reflect on their results. This model, which carries some risks (results are unknown), has been well-received by students, promotes student engagement, and promotes confidence in laboratory science, experiments and research (Weaver et al., 2008; Birkett, 2009). In the biological sciences, these classes are typically reserved for middle or upper division courses, but the largest impact in terms of student engagement, the development of scientific thinking, and the potential for extensive research experiences during college will come from research experiences beginning in the first year.

Introductory biology students at Purdue University are required to complete a one-semester laboratory course that is independent from their introductory biology lecture courses. This course is designed to provide a rigorous treatment of the laboratory skills that will allow students to be successful in their future laboratory courses within the Biology major. While the students are instructed on important laboratory skills, they are not exposed to their application within an actual experimental context. This separation of the skills from their applied use in experimental science is not in line with the current recommendations for undergraduate education which calls for an engagement of students in the process of research as one of six core competencies (Vision and Change, 2011).

In an effort to introduce students to the nature of science early in their college careers, we have sought to increase student excitement and engagement in biology by exposing them to the culture of discovery that incorporates novel research projects into their introductory biology laboratory experience. As part of an NSF-funded project, we have been adapting the CASPiE (Center for Authentic Science Practice in Education) model (Weaver et al., 2006; Weaver et al. 2008) for science laboratories pioneered in the Department of Chemistry at Purdue University to our life sciences curriculum (Bio-CASPiE). In this model, instructional staff work with research faculty to identify novel research projects to use as the basis for teaching students not only the fundamentals of laboratory work (i.e., basic lab skills), but also skills used by scientists as researchers.

In the Fall 2010 offering of Bio-CASPiE, student research projects were driven by research questions about neuroanatomical changes in auditory forebrain structures in an animal model of dyslexia and an animal model of aging. Both of these research models are attractive because they have broad, multifaceted possibilities for research. In addition, the projects have clear “big-picture” problems that the students are working towards rather than perceiving that they are pursuing more esoteric basic research. The auditory thalamus, or medial geniculate body (MGB), is the main sensory input to the auditory cortex, and the MGB can be subdivided further into multiple subdivisions that project to primary or non-primary auditory cortex (Winer, 1992). Despite its pivotal position in the auditory pathway, the role of the MGB in controlling the form and content of neural representations to auditory cortex is often overlooked. Because of this, the MGB and auditory cortex together are fertile ground for the research questions discussed in the next two paragraphs.

Human dyslexics have reading difficulties, often coupled with auditory and other sensory deficits (Fitch et al., 1994; Goswami et al., 2002; Shaywitz and Shaywitz, 2005). Auditory and reading deficits observed in people with dyslexia have been correlated with anatomical abnormalities in the auditory thalamus and cortex (Galaburda et al., 1985, 1994; Livingstone et al., 1991; Escabi et al., 2007). Specifically, the brains of dyslexics often contain cortical malformations known as microgyria that are correlated with a decrease in the number of large cells in the auditory thalamus (Galaburda et al., 1994; de Vasconcelos Hage et al., 2006). Similar cortical malformations can be induced experimentally in rats by placing a freezing probe on the skull overlying somatosensory cortex in early postnatal rats (Fitch et al., 1994; Herman et al., 1997; Rosen et al., 2006; Escabi et al., 2007). Microgyric rats exhibit anatomical abnormalities in the auditory thalamus and auditory processing deficits similar to dyslexic humans (Herman et al., 1997; Clark et al., 2000; Peiffer et al., 2002; 2004). Therefore, microgyric rats serve as an excellent experimental model to determine the neuronal differences in the auditory thalamus and auditory cortex that contribute to the observed deficits in auditory processing. There remains a major knowledge gap regarding the cellular alterations of neurons in the auditory pathway that would underlie the behavioral deficits.

A similar knowledge gap exists for understanding cellular level changes in the aging auditory system. Auditory deficits are present in a growing population of millions of elderly listeners in the USA alone (NIDCD). Deficits are most evident during situations in which there are competing sounds or when the cues to discriminate sounds are weak (Schneider et al., 2005). Much of the research focus has been on understanding cochlear degeneration, but the mechanisms and consequences of age on central auditory structures are not nearly as well understood despite numerous anatomical changes (Caspary et al., 1990; Tadros et al., 2007). Earlier studies have revealed relatively subtle changes in the neural responses of aged animals that do not seem to capture the extent of behavioral difficulties (Shaddock Palombi et al., 2001; Walton et al., 2002). At the anatomical level, one consistent finding across central auditory regions is that markers of inhibitory GABAergic activity decline with age, either when measuring the enzyme for GABA synthesis (Ling et al., 2005) or GABA itself (Caspary et al., 1990). In fact, in the cerebral cortex, this age-related decrease has been shown to be selective for auditory cortex while adjacent parietal cortex showed no such decline (Ling et al., 2005). Therefore, we will be evaluating age-related changes in neural activity in the context of decreased GABAergic inhibition, though other mechanisms will be considered if they fit the data better and more broadly. By determining how MGB responses have been altered by aging, it will be possible to isolate subcortical changes from cortical changes. Moreover, in terms of evaluating a reduction in GABAA receptors in central auditory nuclei as an underlying cause of functional deficits, the MGB has a GABAergic input pattern that is unique among sensory systems (Peruzzi et al., 1997; Bartlett and Smith, 1999), but the distribution and functionality of those receptors is unknown for the aged or microgyric rat MGB.

MATERIALS AND METHODS

Student populations

Thirteen first semester freshman (nine women and four men) enrolled in the research-based introductory biology laboratory took part in the activities and research. Students self-selected into this class, but needed to indicate their intentions of majoring in biology by being concurrently enrolled in the introductory Biology lecture course and the first semester of introductory inorganic Chemistry. The laboratory class met once a week for four hours for the duration of the semester (15 weeks).

Introductory biology students who were not taking the research-based, Bio-CASPiE course formed the comparison population for pre- and post-semester attitudinal surveys.

All data from students have been collected and analyzed in accordance with protocols approved by the Institutional Review Board at Purdue University.

Individual and team assessment

Students were grouped into teams of 3–4 students on the first day of class by the instructor based on responses to online personality and learning styles inventories (Myers-Briggs [http://www.humanmetrics.com/cgi-win/JTypes2.asp] and VARK [http://www.vark-learn.com/english/index.asp]). An effort was made to group together students of varying personalities and learning styles. The students worked with their team throughout the duration of the semester. To facilitate good teamwork and to identify potential problems within groups, the students performed a self and peer evaluation within their teams three times during the semester. This allowed the students to thoughtfully reflect on their performance within the team as well as their teammates’ performances. Students gave themselves a score of 1–4 in the categories of: preparation, technical proficiency, and teamwork (see supplemental material). Each person received a score out of 12 points which was the average of all of scores given for each evaluation.

Assessment by the instructor of student performance in the class took two forms: standard classroom assessments and assignments modeled after scholarly activity. Standard classroom assessments included weekly laboratory notebook checks, weekly quizzes on background information for the week, and participation in the lab activities. Scholarly activity-based assessments included three guided discussions of primary and secondary research literature, three lab reports written in scientific paper format, a research proposal worksheet, a chalk talk research report of preliminary data, and a public poster presentation at the end of the semester (Figure 1) that was attended by Biology faculty and graduate students. Students were evaluated individually on all of the assessments except the research proposal, chalk talk, and poster design and presentation.

Figure 1

Scholarly activity-based assessments. Similar colors indicate how a general scholarly activity was realized as specific class activities.

Organization of the semester

The semester was roughly divided into thirds with the first five weeks being devoted to skills and knowledge building exercises and activities, the middle seven weeks being devoted to the independent research projects, and the last three weeks being devoted to data analysis and presentation.

During the skills building weeks, students worked individually and with their teams to acclimate to working in a lab and to develop the necessary lab skill set and background knowledge to proceed into the research portion of the semester. Each week the students were provided with a laboratory manual, written by the instructor, which included background information relevant to the topic of the course and with specific details for the experiments and procedures being used that week.

On the first day of class students performed sheep brain dissections to become familiar with basic neuroanatomy, with emphasis on the location of the auditory brain structures (auditory thalamus and auditory cortex) that they would be analyzing in their projects. In addition, students started working with bright field microscopy to learn the basics of operation and function. The students practiced viewing and making observations of prepared slides of brain tissue, using an eyepiece graticule and stage micrometer to measure neuronal soma diameters, etc. The second week of class moved forward to the basics of image analysis using ImageJ software (http://rsbweb.nih.gov/ij/) (gridding and random selection of analysis fields across several brain sections, manual and automated cell counts, measuring neuronal soma diameters, and optical density measurements). The data generated from neuronal cell counts and measurement was used to teach the students about basic descriptive and inferential statistics. During weeks 3–5, students built towards being able to understand and perform immunohistochemical experiments. Students learned how to properly use a micropipetter and to perform dilutions with distilled water and food coloring. A blood typing activity with simulated blood (Wards Scientific) provided the students with an opportunity to visually observe antibody-antigen interactions and to design and test hypotheses about these interactions with a quick and reliable system. Finally, students performed a Nissl stain and immunohistochemical stain of rat brain sections (VGLuT2 and calbindin staining, see methods below).

With the background knowledge and skills acquired in the first part of the semester, student teams were asked to formulate hypotheses and design an experiment to investigate the effects of either aging or an induced cortical malformation (microgyria) on neuron number and/or the distribution of some neuronal proteins in the rat auditory thalamus or primary auditory cortex. Students were provided with previously sectioned rat brain tissue mounted onto microscope slides and were to use combined Nissl staining and immunohistochemistry in their study. They had access to the following antibodies: GAD 65/67, VGluT2, calbindin, GABAA receptor (α 1 subunit). Students were given six weeks to perform their staining twice and analyze the results. Students were kept blind to the identity of the tissue (young vs. old or operated vs. sham) until all analyses were complete. Drawing upon their experience during the skills building weeks and with continued guidance from the instructors, students performed all of the tissue processing and data analysis themselves

In addition to the development of laboratory techniques and the planning and execution of experiments, students were given several opportunities to organize, analyze, and summarize their results and present them in written and oral formats.

Tissue processing and image acquisition

All animals were handled and used in accordance with Purdue Animal Use and Care Committee (PACUC) guidelines. Students worked with rat brain tissue that had been harvested and prepared for sectioning in the Bartlett laboratory. Briefly, the rats were euthanized with Euthasol and perfused transcardially with phosphate buffered saline (PBS) followed by 4% paraformaldehyde. The brains were removed and cryoprotected with 30% sucrose in PBS. Brains were blocked into hemispheres, and the spinal cord and olfactory bulbs were removed. Brains were then embedded in OCT embedding medium (Ted Pella). The auditory thalamus and primary auditory cortex were frozen sectioned by the instructors at 30 microns, mounted directly onto slides (Superfrost Plus microscope slides, Fisher Scientific), and stored frozen until needed (Park and Cunningham, 2007).

Students performed Nissl staining and most steps for immunohistochemistry during the course of one laboratory period. Students prepared all of the reagents used for the Nissl stain. Nissl staining was performed in EasyDip slide staining system (Electron Microscopy Sciences) and followed standard procedure with demyelination, staining, destaining, and dehydration steps (Paul et al.,1997; Cold Spring Harbor protocols).

Immunohistochemical staining consisted of a primary antibody with the ABC detection method with a DAB (diaminobenzidine) substrate (Vector Laboratories) following similar procedures to those used for free floating brain sections (Graziano et al., 2008). Students prepared all reagents and determined and performed all dilutions necessary for each step along the process on their laboratory day. In a previous laboratory period, each student group had prepared a 10X PBS stock solution that was diluted to 1X for use in the immunostaining procedure (final concentrations in mM: 154 NaCl, 6.6 KH2PO4, 1 NaH2PO4·H2O, pH 7.4). On the day preceding the laboratory period, the instructional staff initiated the staining process by blocking the tissue (5% normal goat serum in PBS + 0.25% TritonX-100) and incubating the sections in the appropriate dilution of primary antibody (overnight at 20 degrees C) in slide processing chambers (Ted Pella). Primary antibodies and their dilutions in blocking solution were: Calbindin 1:500 (mouse, Sigma), GAD65/67 1:1000 (rabbit, Millipore), VGluT2 1:500 (guinea pig, Millipore), GABAA receptor α 1 subunit 1:1000 (rabbit, Millipore). The primary antibody was omitted from the blocking solution on one slide for each group and served as a negative control for antibody specificity.

The students picked up the staining process the next day with washing off the primary antibody. A biotinylated secondary antibody against the appropriate species was used (1:200 dilution; Vector Laboratories). Following incubation in secondary antibody and washes in PBS, the ABC method was used to visualize the antibody localization (Vectastain Elite ABC kit, Vector Laboratories) with DAB (diaminobenzidine) as the substrate (Vector DAB substrate kit, Vector Laboratories) for the peroxidase enzyme which produced a brown reaction product. Students empirically determined the appropriate reaction times for this step (7–10 minutes for the tissue processed directly on slides). The stained tissue was taken through a dehydration series by the instructional staff. Both Nissl stained and antibody-labeled tissue were coverslipped using Permount mounting medium (Fisher Scientific).

DAB was chosen as the peroxidase substrate for visualization because we wanted to be able to easily compare our results with those published by others in the field. Using DAB as the chromogen as used in the manner described here (with amplification) has limitations for quantitative measurements of antigen numbers (Matkowskyj et al., 2003). However, we are interested in relative optical densities and our stained structures had average pixel intensities of <100 on an 8 bit scale (0 is black and 255 is white) which we feel is still in the linear range (See supplemental Figure 1 for an example histogram of pixel intensity distribution). In addition, DAB is a suspected carcinogen. Students were instructed on the proper handling of the DAB and the instructors removed all DAB waste and neutralized it (3% potassium permanganate and 2% sodium carbonate) before disposal. All work with DAB was carried out with the use of gloves and the slides were processed on top of disposable underpads (Med Vet International) to prevent contamination of bench surfaces.

Images to be used for analysis were captured by the instructor and students using a compound microscope and digital camera (AMscope Clinic Vet Laboratory Trinocular Microscope 40X-2000X, Model T490B and 8 MegaPixel USB 2.0 Microscope Color Digital Camera, Model MD1800). Digital images were analyzed by the students using ImageJ software and included use of the ‘analyze particles’, ‘cell counter’, ‘histogram’ (gray values of selected area for optical density) and ‘measure and label’ tools and plugins (Ferreira and Rasband, 2010). Students were instructed on how to adjust the brightness and contrast and thresholds of their images, as needed.

Peer-Led Team Learning (PLTL) workshops

Students in the class were divided into two groups that met with a student peer leader for five workshops during the semester. The workshops allowed the students to explore topics and concepts relevant to the activity of research in a setting that was not graded. The peer leaders were upperclassmen Biology majors chosen based on their knowledge of the basic research topic of the class and their involvement in the previous offering of a Bio-CASPiE lab (Spring 2010, microbiology). The material covered in the workshops was meant to complement and extend concepts and activities covered in the course. The topics were: how to keep a laboratory notebook, how to write a scientific paper, how to read a research article, and two workshops on ethical conduct in science. Each workshop required minimal outside preparation from the students and consisted of activities that students worked on in pairs and as a whole group (6–7 students). The workshops were all adapted from workshops designed and implemented in CASPiE labs in the chemistry department (Varma-Nelson et al., in preparation).

Evaluation of the effectiveness of the course

Anonymous responses to the online institutional end of the semester student evaluations were gathered to determine student satisfaction with the course and evaluate student feedback in the free response area within the evaluation.

In addition, an attitudinal survey consisting of 37, Likert type items regarding student experiences in their previous Biology laboratory classes was administered to Bio-CASPiE and non-Bio-CASPiE students before their participation in their respective biology laboratory classes. A similar survey consisting of the same items was administered to the students at the end of their fall 2010 biology laboratory experiences. Response categories for the items on the attitudinal survey were: 6 = strongly agree, 5 = agree, 4 = barely agree, 3 = barely disagree, 2 = disagree, 1 = strongly disagree. Thirty-nine students responded to the pre-survey and 32 students responded to the post-survey. However, there were only 12 successfully matched pre-post pairs (four Bio-CASPiE and eight non-Bio-CASPiE). Due to the small sample size, ordinal structure of response categories, and non-normal distribution of the data, a non-parametric statistical method was employed to examine differences between the two groups. Specifically, Mann-Whitney U analyses were conducted to evaluate group differences in the median scores on the pre- and post-participation survey items.

RESULTS

Direct application of basic skills and acquisition of research-related skills

Similar to laboratory courses at other large universities, the traditional introductory biology laboratory course offered to freshmen at Purdue University focuses on the acquisition of a skill set that will be useful to students as they progress through their coursework as Biology majors. The research-based approach to introductory biology laboratories aims to provide students with the same necessary skills, but with an immediate application of the skills to a research project that the students are engaged in. These skills include the proper use of a balance, making molar and percent solutions, proper micropipetting technique, performing dilutions, using a pH meter, using a compound microscope and performing descriptive and inferential statistics. The learning of these skills occurs during the skills-building portion of the semester before the students embark on their research projects.

Because of the research theme of this laboratory course, students in the Bio-CASPiE lab acquire some research-related skills that are not explicitly taught in the traditional laboratory course (Table 1).

Table 1

Research-related skills acquired by students in introductory biology laboratory courses

Student-directed experiments and analysis

One of the primary goals of this research-based approach to introductory biology laboratory courses is to increase student engagement in the coursework. In an effort to facilitate this, throughout the semester, students were given the opportunity to design several aspects of their experiments and data analysis, discovering the limitations of their design and being able to refine their approach during the semester. The first opportunity for this “design and refine” approach came with the data analysis for their first lab report. They had been given basic instruction and guidance for making neuron counts and measuring neuron soma diameters from Nissl stained brain sections from the instructors, but each team was responsible for coming up with the specific methods that were to be used by every member of their group. Each team member was responsible for making measurements that would be pooled by the group to form their data set to report on. In the lab report the students were asked to identify potential limitations of their analysis and propose ideas for improving their approach. Two common themes emerged from these reflections: 1) imperfect standardization of analysis techniques across all team members (criteria for distinguishing neurons from glia in Nissl stains and uniform measuring of least and greatest neuron diameters) and 2) small sample size of neurons counted/area and number of brain sections analyzed. Based on these reflections, the students were asked to reconvene with their teammates and refine their data analysis procedure to create a standardized, reproducible technique that could be employed. They were then asked to re-write their lab reports using the new analysis procedure that they designed.

The second area where students were responsible for the direction of their work in the class came during the independent experiment section of the course. Student teams completed an experiment proposal worksheet written by the instructor (supplemental material) to assist them in the experimental design process. The worksheet consisted of the following elements: 1) Relevant background information, 2) Statement of their research question, 3) Rationale and hypothesis to be tested, and 4) the basic experimental protocol to test the hypothesis (what will be quantified, the test conditions, controls, number of repetitions, and the type of statistical analysis to be used). Each team met with the instructor to discuss their proposal and to converge onto a set of experiments that could be finished within the rest of the semester. There were four teams in the class and the projects topics were: 1) Neuron numbers in the medial geniculate body in young and aged rats, 2) Distribution of GABAA α1 receptor subunit in the auditory cortex of young and aged rats, 3) Distribution of VGLuT2 and GAD65/67 in the medial geniculate body of young and aged rats, and 4) Distribution of calbindin in the medial geniculate body and primary auditory cortex in microgyric rats.

Products from the research projects

Research teams worked with instructors in the course to design and carry out analysis of Nissl and immunostained sections of rat auditory brain structures. The types of quantitation included neuron cell body counts and/or optical density measurements of antibody labeling in their experimental and control rate brain tissue. These data formed the basis for two lab reports, a chalk talk on their preliminary findings (supplementary material), and the final poster at the end of the semester. Through these assignments students could explore various means of organizing their data for visual presentation (tables, bar graphs, images of stained brain tissue). Examples of student-generated figures from each of the projects are shown in Figures 3–6.

Figure 3

Example of student quantitation and presentation of immunohistochemical data. Example images from immunostaining of auditory cortex with the GABAA α1 subunit. The bar graph is a summary of average ± SD of the relative optical density of...

Figure 6

Example of student quantitation and presentation of calbindin-positive neuronal counts in the MGB in microgyric and sham-operated rats. The top image shows calbindin-stained MGB in a control rat gridded for cell count analysis. It is accompanied by the...

The individual lab reports and team chalk talk helped students organize their data, evaluate it, and provided check-points for feedback from the instructor. At the end of the semester, each team organized their project and its data into a poster that was presented to the public.

Evaluation of the effectiveness of the research-based introductory biology laboratory model

Course evaluations of students in the CASPiE course and attitudinal pre/post surveys of CASPiE and non-CASPiE students formed the basis of evaluation of the CASPiE model of introductory biology laboratory courses. Standard course evaluations completed by the CASPiE students at the end of the semester indicate that the students felt that the course was an excellent learning experience (Table 2).

Table 2

Student responses to institutional course evaluations. [E] Excellent=5 [G] Good=4 [F] Fair=3 [P] Poor=2 [VP] Very Poor

Student feedback in the free response portion of the course evaluations was generally very positive and included the following comments: “Overall, the class was very well organized. I didn’t feel overwhelmed with information, but was still able to start the semester with no research experience and end the semester having completed a research project that was interesting and that I can be proud of. I also liked starting the first day of class by dissecting sheep brains.” and “Excellent class and excellent professor that put A LOT of time and effort into the course. I wish I would have worked harder myself throughout the semester so that I would have got even more from it.” Most constructive criticisms of the course were related to specific activities in it and not the model of the class itself. One student remarked, referring to the first day of class: “In the future, especially for freshmen taking this class I think it would be really nice just to get to talk before the first class, because it was a very intimidating experience.”

Introductory biology students enrolled in the Bio-CASPiE lab and students enrolled in the traditional, skills- based lab were asked to complete an online survey at the beginning and the end of the semester to evaluate their attitudes about their experiences in their most recent laboratory courses. The pre-semester survey probed their experiences in laboratory courses prior to that semester and the post-semester survey probed their experiences in the laboratory courses that they were enrolled in during the fall of 2010 using a Likert scale for responses. Only data from students who completed the pre and the post semester surveys (matched pairs) were included in the analysis of the data. There were no significant differences in the responses on the pre-semester survey between CASPiE and non-CASPiE students, suggesting that the two populations were not significantly different with respect to their past laboratory course experiences (data not shown). However, there was a significant difference between CASPiE and non-CASPiE students for several response categories on the post semester survey (Table 3).

Table 3

Group differences in post-participation survey responses.

DISCUSSION

Student engagement in research

Undergraduate research experiences have increasingly become among the minimum necessary requirements for admission into graduate schools (PhD programs as well as medical and veterinary school). However, these experiences are varied depending on the expectations of both the research mentor and the student. In a review of faculty-mentored research experiences, Wilson et al. (2011), identified different categories of experiences that depend on the perceived goal of the experience as viewed by the research faculty: 1) retention or selection 2) research exposure 3) advanced learning or de facto streaming 4) developing scientific research skills 5) thinking like a researcher and 6) entry into the research culture. These categories lie along a continuum with respect to the role of the student in the research project and process ranging from largely passive (1 and 2) to a more active participant (4–6). As a result of these varied goals, the degree of active involvement of the students in the design and execution of the experiments and enjoyment of the experiences and learning is also highly variable (Howitt et al., 2010).

The advantage of the Bio-CASPiE laboratories is that students are provided with methodical and consistent guidance by the instructors while still giving the students the freedom to direct some aspects of their experiments and data analysis. This structured experience with novel research has thus far been a positive experience for the students, instructors and research faculty (see below). This is a departure from most of their experiences prior to the class where labs have consisted of exercises with known outcomes and can be unsettling to some students. Students are encouraged to embrace the unknown as an exciting challenge. A further goal of the present study is to present a developed and assessed course so that faculty are also encouraged to embrace the unknown in their courses as an exciting challenge rather than a daunting hurdle.

Research quality data

The students in the neuroanatomy Bio-CASPiE course were able to generate good quality preliminary data for the collaborating research faculty along several different lines of inquiry within the scope of the general research questions developed for the class. Some of the brains used in the undergraduate research were from animals whose electrophysiological responses to sound were recorded (Parthasarathy et al., 2010). Many of these data will be followed up in the Bartlett lab, with summer research students (directed by Dr. Gardner), and in the next offering of the neuroanatomy course in the fall of 2011. In addition, we are including some of the CASPiE student data in an abstract for the upcoming Society for Neuroscience meeting, including the students in the trip to the meeting and presentation of the data.

As is the nature of any research program, the projects done by students in all of the Bio-CASPiE classes will evolve as data are generated and new hypotheses are formulated. We have already begun to experience the evolving nature of the projects and progress made by the teams of undergraduate researchers this spring with the second offering of our microbiology Bio-CASPiE course. The data gathered by students in the spring of 2010 has been pursued in the lab of the researcher (LN Csonka) and by the new crop of CASPiE students. The current students have been inspired by the fact that they are further analyzing and contributing to projects initiated by their fellow Biology majors in the previous class!

Limitations and future directions

Our initial experiences have been very positive and the students involved have expressed their enthusiasm for the course in course evaluations and indicated that the course had a positive impact on their attitudes surrounding laboratory science. However, there are limitations of the approach we have taken with respect to its broad adoption in undergraduate curricula. This approach is time-intensive in its development and implementation during the semester and requires dedicated instructional staff. The staff needs to familiarize themselves with the research topics and techniques as well as consult with the collaborating research faculty before and during the semester. In this sense, it helps to have TAs whose research is similar to that being used in the course. Encouraging research faculty and their postdoctoral fellows, graduate students, or experienced undergraduates to be directly involved in delivering the class would alleviate this. This would provide an excellent teaching experience for the postdoctoral students and/or graduate students because they would have a greater connection to the work done in the class and this would likely translate into their investment in quality instruction. Repeating the course with similar topics to those previously offered allows the instructional staff to improve on past courses, see the projects evolve, and reduce the time and effort needed to start a brand new topic.

We have chosen ambitious projects for our students to become involved in, but feel that scaled-back versions would be highly valuable experiences for the students and reduce the cost of the lab course/student. For example, an approach similar to the one used by Birkett (2009) in which students had access to previously-stained brain tissue could be used. This would decrease the cost of the course as the expensive reagents, like antibodies and tissue would be incurred by the research faculty within the normal course of their research program. Students could perform some staining of other cell types or a small amount of brain tissue, to get the practical experience to connect them to the research. In many cases, there may be data that a faculty member has fully or partially collected but has not had the time to complete the data collection and analysis.

We have thus far only had the opportunity to evaluate the near-term impact that taking this research-based course has had on student learning and attitudes about laboratory science. We are currently following the alumni of these classes to determine what lasting impact this experience has had on their attitudes about laboratory science, the likelihood of seeking out other undergraduate research experiences, their retention within the biology major, their performance in courses within the biology major, and their career goals and plans.

Figure 4

Example of student quantitation and presentation of neuron counting data in the MGB in young and aged animals. On the left is an example image of Nissl stained ventral MGB with a 200x200 micron grid superimposed on it. The students randomly selected a...

Figure 5

Example of student quantitation and presentation of VGluT2 and GAD 65/67 axon terminals with student-written figure legends. Students counted puncta from randomly-selected grid squares across multiple tissue sections to get average axon terminal densities....

Categories: 1

0 Replies to “Biology Coursework Skill Deficit”

Leave a comment

L'indirizzo email non verrà pubblicato. I campi obbligatori sono contrassegnati *