20.309:Lab Report Guidelines

From Course Wiki
Jump to: navigation, search
20.309: Biological Instrumentation and Measurement

ImageBar 774.jpg




Convincing (XKCD #833)


Lab report guidelines

The purpose of your report is to precisely and clearly communicate:

  • what physical phenomena you examined in the lab;
  • the means and methods you utilized;
  • your results (which must include a measure of significance or uncertainty);
  • the process you used to analyze data; and
  • what you learned.

The report should be brief. Consider each pixel you choose to darken on the page in the context of the information it conveys. Ideally (but not practically), erasing a single pixel would diminish the report. This does not mean you should use a small font.

Imagine you are writing the report for a classmate. The intended audience has a deep understanding of the subject. (Perhaps you need to imagine a different classmate.) As such, lengthy derivations and explanations are not necessary. The reader would very much like to understand what you did in the lab, what sort of problems you ran into, how you analyzed your data, how significant error sources may have affected your results, and what you concluded.

Your report does not have to be written like a formal scientific paper. Use whatever organization works for you so long as it includes all of the information.

  • Bullet points are acceptable.
    • Bullet points offer a more concise way to present some kinds of information than flowing text.
    • Hierarchical organization may make bulleted lists more readable.
    • Use numbered lists only in cases where the order of the information is critical or you must refer back to the information by index.
  • Brevity is not an excuse for imprecise, incomplete, or unclear communication.
  • Follow the content and style guidelines below.

Explain the apparatus and procedure

Some lab exercises involve significant design and construction work. Others utilize complex systems that were built by others. In either case, the apparatus you used in the lab must be documented in your report. Block diagrams, schematics, photographs, text descriptions, and citations of sources are all excellent ways to describe an apparatus.

Design documentation should be sufficiently detailed to enable another student to build a functionally similar piece of equipment. Include component values and reference designators (Rl, Cf, L1, etc…) on schematics. Explain why you chose the values you did. If you chose to implement a portion of your apparatus differently than the lab manual suggested, explain why.

Detailed documentation is available for the predesigned pieces of equipment in the lab (such as the AFM and the optical trap). In these cases, refer to the documentation and discuss aspects of the machine that are essential to understanding your method and analysis. A simple block diagram and a few bullet points will meet the requirements.

Once you have documented the apparatus, explain how you used it and what measurements you took.

Explain your analysis

Your report must explain how you analyzed your data. It is unnecessary to reproduce each step of every calculation. Include enough detail to make the process clear. If you wrote (or used) computer code, explain how the code works (possibly using bullet points) and include the source code in an appendix.

Compare your results with expected values from theoretical models, literature or other lab groups.

Report numerical results properly

All numerical results should be reported with an appropriate number of significant figures and an associated uncertainty or significance level. There should be no ambiguity about the units of all numbers reported. Use plots or tables to report multiple results efficiently.

Datasets must include appropriate summary statistics and an appropriate measure of variability, such as a range, standard deviation, or standard error. Include the sample size. Use the abbreviation "s.d." for standard deviation and "s.e.m." for standard error after the "±". For example: 1.21 ± 0.03 GW (±s.e.m., n=42). Round uncertain quantities to the same decimal place as the uncertainty. Uncertainty is typically reported with one or two significant figures.

Numbers should not have a large number of leading zeros. Use scientific or engineering notation or appropriate units. Acceptable: 3x10-5 m, 30x10-6 m, and 30 μm. Unacceptable: 0.00003 m. There should be a space between the number and the unit symbol.

For each numerical result, discuss significant sources of error and indicate whether the resulting error caused a systematic or random distortion of the results. To the degree possible, discuss the sign and magnitude of the error and how the effect of the error source could have been mitigated. A significant portion of the lab report grade depends on your error analysis. If you are uncertain how to calculate or report the uncertainty of a numerical result, ask an instructor for advice.

Say what would you do differently next time

The instruments and procedures you will use in the lab are not perfect. Explain what factors limited your measurements. Discuss potential approaches for improving results.

Answer questions posed in the lab manual

Ensure that your report addresses all the questions in the lab manual.

Present figures, plots, tables, and images properly

  • Figures must be numbered, descriptively captioned, and discussed in the text.
  • Plots must have a brief, informative title. Axes must be labeled with a description, including units.
    • The font size used in plots must be large enough to be legible.
  • Images must include a scale bar. (Pictures of your apparatus are an exception.)
  • Chartjunk is forbidden.
  • Use tables appropriately.
  • Try not to use screen captures.

Do not present every piece of raw data you gathered. Include only things that are important, in a clear and concise format. If for some reason you have a large body of extremely interesting data to present, put it in an appendix and refer to it in your report. If you have a large body of uninteresting data, leave it out.

Write tight prose

  • Avoid the first person.
  • Use active verbs.
  • Eliminate unnecessary words.

It is common practice in scientific writing to avoid using the pronouns I and we. This is changing slowly; however, it is an excellent habit to minimize the first person in scientific writing. The author of a first person sentence in a scientific paper has frequently misidentified the true subject of a sentence. Consider the following 4 versions of a sentence:

  1. I ran the experiment three times and calculated an average power of 1.21 GW.
  2. The experiment was run three times and an average power of 1.21 GW was calculated.
  3. Three experimental runs yielded a calculated average power of 1.21 GW.
  4. Three experimental runs yielded an average power of 1.21 GW.

In this example, substituting the subject “The experiment” for “I” eliminated the first person pronoun in sentence number 1, resulting in somewhat preferable sentence 2. The second sentence, though, is undesirable because it contains the passive verb forms "was run" and "was calculated." Passive voice is common in scientific writing. That does not mean you are required repeat the offense.

It is impossible to completely eliminate passive sentences from scientific writing. But you can try. Try to limit your use of the passive voice to no more than tweny percent of the verbs in your report.

Sentence 3 is preferable to number 2 because it uses an active verb. But it still could be tighter. Average values are necesarily calculated. Unless required to differentiate between several possible averages, remove the redundant verbiage.

Use terms precisely

Technical terms such as error, uncertainty, resolution, bandwidth, sensitivity, and responsitivity have precise meanings. Use them with utmost exactitude.

Computer code

Include computer code that you developed in appendices at the end of the document. In the text, provide an outline of each function or script you used. Indicate the source of any code you did not write yourself.

Write your own report

In 20.309, you will work collaboratively with other students. The report you submit must consist entirely of work of done by members of your lab group. Any words, data, images, code, or other intellectual property you take from somebody else must be appropriately cited. Take special care to cite computer code and datasets that came from other sources.

Plagiarism is unacceptable.

Policy on data sharing

You are encouraged to compare results with other groups. Clearly indicate the source of any data, code, or results generated by others.

Submitting reports

  • One member from each group should submit a single file, in PDF format, to Stellar in advance of the deadline.
  • The filename should be a CamelCased list of the last names of each group member. For example: CrickFranklinWatson.pdf

Grading

Lab reports will be graded on a 10 point scale. An outstanding report:

  • Demonstrates mastery of the theory and practical implementation of the key techniques used in the experiment.
  • Presents results for each of the required experimental procedures
  • Clearly details error sources and uncertainty of results.
  • Draws appropriate conclusions from the data
  • Contains a thorough, correct, and well explained analysis
  • Addresses shortcomings of the experimental procedure and problems that arose during experimentation
  • Acknowledges everything that is not original work with an appropriate citation
  • Is well communicated