Labs (and the Not So Obvious Use of Labs
by Non Scientists)
By Ken Cheney
I am eager to learn from you, so please
e-mail me your opinions!
the experiment smells it’s chemistry, if it wiggles it’s
biology, if it fails it’s physics."
What are Labs?
I consider labs to be any learning situation where the
student is actively involved, in contrast to a pure lecture situation
where the student passively adsorbs content provided by the teacher.
"Lectures" where students are actively asking and answering
questions will even come under this definition, although I won't
discuss them much here (See "Lecturing"). I will
consider activities from sports to post doc. research!
Unfortunately, I have mostly discussed “experiments”
with equipment. Well, that is what I know best! I believe there
is a lot of overlap with training (e.g. planning, safety, practice,
. . ), so I hope this will be of help to those doing “labs”
that are not experimental.
Who am I talking to?
I visualize several possible audiences. You might
be an experienced teacher new to labs, someone experienced in
labs but new to teaching, someone new to both labs and teaching,
or (conceivably) someone experienced in both labs and teaching
who is just hopeful of finding some ideas new to them. I
have strong opinions as to what "should" be taught in
science labs in particular and in lab situations in general; I
hope I will not offend all my readers all of the time. If
you are offended, I hope you will carefully consider why I am
wrong (and let me know!)
As you will see, I feel labs are best taught by those experienced
in doing the topic of the lab. Of course in the real world
this is not always (usually) possible. If you want (or must)
teach labs without practical experience, I hope some of my hints
will prove useful.
Style and Presentation:
Following my own advice about presentation styles, I have
written this essay informally in the first person with liberal
use of personal examples and opinions. The reader should keep
in mind that they are personal opinions; little is certain in
education! My examples are, unfortunately, heavily weighed toward
science. This selection simply reflects my experience, not my
opinion that one field has a monopoly on good or bad practices.
The plan will be to cover the subject in a broad, shallow
survey rather than by a deep focused analysis.
"My background in experiments and
- I’ve taught Physics, Programming, Laser Technicians,
driving (the family business) and gymnastics (three national
champions used routines I taught them).
- I’ve originated or completely redesigned entire
labs for laser technicians, computers for laser technicians
and computer graphics.
- I’ve designed almost fifty new individual labs
for physics classes (never all used at the same time) and many
more for lasers.
- I’ve written manuals for labs for Physics majors,
for Life Science majors, for Liberal Arts majors, for Programming
for Laser Technicians, for Fortran, and for Computer Graphics.
- My professional experience in labs consisted of four
years working in the aerospace industry (directing experiments
in support of electric propulsion for spacecraft etc.) and a
summer at JPL working in support of the solar simulator.
- I'll discuss many aspects of teaching labs, all the
way from surviving Murphy's Law to planning a complete lab course.
"Why do labs? If you understand
them there is no need to do them. If you don’t understand
them there is no point in doing them.”
"Ahhhh, but you just might learn by doing!!!!!!!!!"
What’s different about labs?
What can be done in labs better than in lectures? Much
research answers “everything”, but here are a few
specifics. Specifics for science labs are listed below!
- Teach professional attitudes
- Personal interaction with students. Learn their majors,
past experience, plans for the future, . .
- Learning to distrust data!! And theory!!
- How to do sanity checks on data and theory
- How to deal with imperfection!
- Lab reports
- Experimental design
- Data analyses
- Students can develop confidence in their own ideas and
- Your students can learn many more real world tools than
they could just listening to lectures.
- Morals for any type of lab:
- Don’t give up!
- Keep Thinking!
- Do your best, and be proud of it.
- Show class!
- Keep your cool!
What’s fun about labs?
- You can interact with students.
- You can try different approaches to teaching and experimenting.
Lectures tend to leave you lecturing!
- With much more time in a block, you can try class activities
not practical in a one hour lecture section.
- Students can be creative.
- Students can see their own ideas carried out!
- Students can watch as their skills develop noticeably.
Labs are taught in many disciplines and for many reasons.
There might seem to be little in common between football practice
and labs being done by a biology post doc. However these labs
may have more in common than physics labs being carried out in adjacent
rooms for Liberal Art students and for Physics majors. Let
me discuss the objectives of various types of labs:
1. Training (vs. Education):
Perhaps the majority of "labs" are training
in the sense that they are intended to install specific skills,
either physical or mental. This is in contrast to "education"
where developing generalized skills and thought processes is
Naturally, there is considerable overlap. A welder
should be able to generalize a bit to different pipe sizes,
etc., while even a very sophisticated chemist needs good habits
while working with toxic chemicals.
The teacher of a laboratory course probably should carefully
consider what they desire to teach in the way of education or
training. It may be best to take a break from sophisticated
considerations occasionally to master some of the tools of the
experimentalist trade. In the case of physics, experimentalists
need to know how to use oscilloscopes, signal generators, interface
computers, build simple amplifiers, etc. These topics
aren't (for this purpose) examples of electricity and magnetism
but are simply tools of the trade.
On the other hand, training can be made more valuable
by "education". I suspect a welding instructor
might occasionally take a break from the skills of welding to
contrast other methods of fastening: gluing, bolts, rivets,
. . . (not to mention the various kinds of welding: arc, gas,
electron beam, . . .). This diversion is not to train experts
in other techniques but to give the students a wider view so
they may recognize when some other technique (or combination
of techniques) is more appropriate - education!
2. Doing Research
In these labs the students are actually doing research
to find out something new. The research may not look fundamental
to someone experienced in the field, but it is designed to be
near the limits of the capabilities of the students involved.
A gymnast may be investigating how reliably she can do a new
skill in a routine, or physics students may be investigating
how wildly real friction differs from the textbook variety,
but in any case they are exploring something unknown to them
and, perhaps, to anyone. This type of lab exposes
students to the joys and frustrations of the real world.
There is no guarantee of success, but what is learned is something
the student can regard as their own achievement.
It might be thought that everything accessible to students
was well understood by "the ancients," but once one
is looking for them, there seem to be an unlimited number of
small but interesting problems that are not well understood.
Of course research in a freshman class may not set these problems
to rest, but here it is the journey not the destination that
Of course everyone (students and teacher) must take the
results with a grain of salt. You probably wouldn't design
a skyscraper based on the tests of welding techniques made in
a welding class. However the student welders who designed
and executed the tests will have a much greater appreciation
of what may or may not work as they see their welds fail under
test loads. Conversely the teacher shouldn't expect polished
results (lack of time, materials, knowledge, . . ).
It’s the process that counts, not the product!
3. Historical Labs
These labs redo classical experiments, either following
as closely as practical in the footsteps of the originator or
using modern methods to achieve the same type of result.
Such reenactment can be very satisfying to students. I,
and many other physicists, fondly recall doing the Milliken
Oil Drop Experiment (the first experiment to determine the charge
on an electron). Our experiments were not easy, clear
cut, or accurate. But, we had actually dealt with one
of the fundamental experiments in all of physics!
These labs often can't realistically be considered training
or education, but they are very important to the culture of
4. Time Wasters (keeping the kids off the streets):
“Teach something!” Perhaps the greatest
impediment to teaching is that students have learned from experience
that most of school is simply useless stuff designed to keep
them busy. Figure out something really educational
for the students to be doing! Student designed,
researched, and executed labs are bound to be educational, but
perhaps hard on the teacher!
5. Proving Theory:
For a scientist this is the most frightening type of
lab of all! Why? To claim that you are going to “prove”
or even “verify” a theory is to strike at the very
heart of science!
Theories cannot be proved! Theories can
be checked or disproved. The essence of science is that theories
(if they are to be taken seriously) must make predictions that
can be checked (be falsifiable). The theories then should
be checked. Finally, most important, the theories
that made wrong predictions are discarded, no matter how fond
we are of them.
Theories surviving many attempts to disprove them are
treated with more and more respect and used with more and more
confidence. Does this mean that the theories are "true"?
No, it just means they make predictions we can't disprove with
our present accuracy of measurements. In some fields,
measurements are so crude that virtually any theory can be twisted
to match the data (take theories of education for example).
In contrast, it is known, experimentally, that gravitational
and inertial mass are the same to 1 part in 100000000000000000000
or so. But it is not known that they are
the same. Physics is strewn with excellent theories (Newton's
Mechanics, light as a wave, heat as a fluid, . . .) that work
fine for most purposes but that fail in "extreme"
The problem with "proof" is that we cannot
try all possibilities, and we cannot measure to an infinite
number of significant figures. Therefore we can't be sure
that the theory works perfectly in all cases. At
best, we can say that for the conditions we tested, the results
were within some percent of theory. With a bit of statistics
we can say what the probabilities are that this difference is
significant (i.e. that our results probably disprove theory
or that the difference between our results and theory is likely
due to random errors).
To say a lab is to "prove a theory" is to miss
a vital element of science! We can check theory, we can
disprove theory, we can not prove theory!
6. Hands on for the experience:
A type of lab that is very useful despite the fact that
it is neither training nor (strictly) educational is the "play
with the equipment and concepts" lab. Most
people find it hard to visualize things that are just described
or even pictured. It generally helps a lot to have a personal
experience with the objects or concepts. More, most people are
delighted to find that they can actually operate an instrument,
plot a graph, do a flip (sort of) even if not very well.
In the future when the student sees or hears of that activity
they will say to themselves "I’ve done that and I
know what its like!"
7. Reinforce theory by using it for real:
This type of lab can be a dull “follow the manual”
experience or an exciting, creative activity.
8. “Follow the manual”:
The manual spells out in detail the steps to follow,
the boxes to put the numbers in, and, perhaps, even the conclusion
to draw. The apparent upsides to this approach are that the
lab looks very organized, the students feel secure, the students
have a good chance of “success”, and the lab is
over quickly. The downside of this plan is the elimination of
almost any chance of the students learning anything.
"If your bubble never bursts you’re not pushing the
9. "Here are the tools and the problem, figure out for
yourselves how to solve the problem”:
The upsides of this approach is that the students stand
a very good chance of actually thinking about the tools (theoretical
and mechanical) they have to work with. Having devised ways
to use these tools the students are almost certain to have gained
a much better appreciation of what can be done. Further, since
a considerable amount of student thought has probably gone into
the experiment, the odds are good that the students will retain
much of their knowledge.
The downsides of this approach are that the lab appears
quite disorganized and that many students will be uncomfortable
as they are not told exactly what to do. This lack of comfort
can lead to many complaints and frustration with the lab. There
is no doubt it is difficult for the teacher to judge precisely
how much help to give to insure a good ratio of success to suicide.
The students can’t be safely left on their
own; they will mostly attempt impossible designs or not plan
at all. You must assign milestones that you have to approve,
and you must regularly consult with the students as to how their
design or experiment is going. At least two weeks are required
generally; the first week (if luck holds) eliminates many impossible
devices leaving (you hope) workable devices for the second week!
Administrators and other teachers often haven’t
a clue as to the point of such labs! It probably
is a good idea to educate supervisors beforehand about the philosophy
behind the lab design and the possibility of failure.
Examples of “figure it out for yourself”
When I was teaching programming I would generally present
a programming tool, arrays say, give an example or two or
their use, and assign the students to write a program about
their interests using the new tool. Students would generally
get into the intricacies of their own problem and do much
deeper work than they would have just solving a problem I
There is a powerful concept in physics called the conservation
of angular momentum. This is familiar to dancers and gymnasts
who spin fast with their arms in close to their body and spin
slowly with the arms outstretched. I have many times assigned
my physics lab the problem of designing an experiment to check
this law. Each group was to invent a different approach,
then actually build and use their apparatus. Students who
had previously displayed little interest in the “canned”
labs suddenly showed quite unexpected powers of creation as
they developed their own strange and wonderful machines.
Not all the machines worked. (They didn’t have to work.
A good try and analysis led to a good grade.) But all the
students received a great grounding in the conservation
of angular momentum and in the perils of experimental design.
Deeper for an extensive list of what can be taught in science
labs, and why such labs are important.
Deeper for thoughts on why labs should be taught by experimentalists.
An ideal lab might have the following properties:
- Surprises: Students will be more interested in
labs that provide surprises. The surprise could be in many forms:
The results of an experiment (seen at once), the results of an
analysis, the discovery by the students that they could do something
they never expected to do, the “Ahhhh” moment when
the students make a connection between theory and everyday life
. . .
- Transparency: The connection between the parts
of the experiment and between the measurements and the conclusions
should be easy to follow. A great experiment that looks like
a black box to the students is useless.
- Extendability: The students should
understand the method and equipment so well that they can apply
the principles to other applications.
- Student contributions to the design of their experiments:
Students will learn much more, and develop more confidence, if
they do part of the experimental design, the more the better.
- Important topics: Are the topics important or just
- Optimum experimental design: Is the method of doing
the experiment the best for achieving goals, such as those listed
here, or is it just the traditional design?
- Involves general-purpose equipment: Are
the tools the student learns about in this experiment going to
be of general use to the student or are they specialized to this
- Teaches tools of the trade: Does this
experiment lend itself to teaching some necessary “tools
of the trade”? These tools may be safety checks, a critical
attitude, oscilloscopes, safety around an imploding vacuum system,
how to fall when you miss a catch in gymnastics, preparing beforehand,
how to read manuals, or any of dozens of skills and attitudes
necessary for success in life.
- Appropriate Difficulty level: This
can easily be a potent source of conflict between ambitious teachers,
students, supervisors, and colleges! Naturally your baseline depends
on your students' abilities and the purpose of the class: Liberal
Arts Physics for Poets, Transfer to Caltech, training journeyman
welders, preparing for MSATs, etc.
- High but attainable standards:My students often
complain: “but this is hard to do”, “…figure
out”, “…debug”, etc. I tell them : No
one will pay them BIG BUCKS for doing easy things. Students
seem to identify very well with this idea once it is pointed out
to them! I’m an advocate of high, layered, standards. Students
will, generally, adapt to the standards expected of them if the
expectations are attainable. If the expectations are low the
students will learn little, be unprepared for future courses or
jobs, and, worse, expect life to continue giving them a free ride.
- Requirements that can be layered: There can be
minimum requirements for everyone (“C”) and suggestions
or requirements for better grades. For example, it may be enough
to take some plausible data and do some analysis to get a C.
To get an A all parts of the write up should be good but, more
important, the results should be examined and any interesting
I tell my students that pointing
out the interesting points is enough, they don’t necessarily
have to explain these points (no one understands everything).
However, I claim that at Caltech these aberrations would be
the start of the experiment, not the end!
Foolproof (low expectation) labs:
The opposite end of the difficulty spectrum is to design
(buy!) labs that cannot fail and require almost no analysis.
This type of lab will meet with little resistance from most students
or supervisors. For labs offered for cultural purposes (Art appreciation,
Physics for Poets) this approach may be quite suitable.
What do you want to teach? How deeply? Even after
the most careful preparation, you should expect to make many modifications
as the result of your experiences with the class, department,
Why are the students taking the course? What do
they need to do with the knowledge they gain? Live a fuller life,
vote intelligently, run an auto repair shop, pass a MSAT, go to
graduate school in your discipline, . . .
What is the background of your students? Right out
of high school, still in high school, returning students, majors
in your discipline, majors in auto shop. . . Students who know
something (i.e. have lived a little) are MUCH easier to teach
than students straight from high school who carefully avoided
actually learning anything!
- What are their native languages?
- Do they understand English? Spoken? Written? How well
can they write?
- Do they have the background necessary for your course?
Math? English skills? Maturity? Motivation? . . .
- How much do the students already know?
- How much must the students know? About the equipment,
about the theory?
- How much would you like the students to know
by the end of the experiment? About the equipment, about the
- Are there “tricks of the trade” that you
can teach here (e.g. laser safety, how to spot skills in gymnastics,
how to unplug a power chord . . .)?
- Are there features of experimental design to be taught
here? e.g. controls, clamps (for safety), avoiding systematic
errors . . .
Murphy never Rests
Check lists: Preparing for a lab – survival and
||What equipment is to be used?
||Can you make the equipment
||What can you think of
that the students can do to make the equipment fail?
||Is there enough equipment for
each student / group?
||Does all the equipment work?
||Are there quirks to the equipment
that you and the students should be aware of?
||Are the manufactures’
||Are detailed instructions or
check lists available for this experiment?
||Can you study good student reports?
||Can you audit an experienced
||Can you consult with an experienced
teacher about quirks of the experiment, good methods to present
the experiment, things the students often get wrong, . . .?
you do during the lab? Beware of Lecturing!!
Students want to get their hands on the equipment. Tell
them just enough so they are safe and can do something then let
them start. After a few minutes they will realize they don’t
know everything (but they will know lots more about the experiment
then they did originally) and will be much more receptive to your
hints and suggestions for further investigation.
Have a minimum of talk, a maximum of
"But what do we do?”
It is disheartening to hear this question just after two hours
of brilliant lecturing on what they should do. Life is better
for you and the students if you feed them the information five
or ten minutes at a time with hands on work in between. Have
micro lectures several times during the lab.
About lecturing: Methods of presenting the material and
involving the class
"Tell, Tell, Tell:" The
advice of the U.S. Army to teachers is "Tell them what
you are going to tell them, Tell them, Tell them what you told
them." Good advice. But, telling
is not as good as showing, and showing is not as good as doing.
of presenting the material and involving the class
Tell what you are going to "teach"
This will help greatly in keeping the
audience oriented. Listeners often have trouble separating the
point of the lecture from the surroundings (i.e. they can’t
tell the steak from the sizzle unless you help them).
Tell why the audience should care!
Everyone pays more attention if they know
the knowledge they are about to acquire will help their health
or happiness (i.e. "Your chances of having cancer will
be reduced by 30% if you follow these food guidelines").
What applications can you include?
The more real world (or philosophical)
applications you can reference, the more likely it is that the
audience will identify with your point.
What are amusing parts?
About grammar: "Piano for sale by
lady with mahogany legs"
Is there some interesting history to
What questions can you ask the class
to force them to actually use the concept?
After outlining the classic plot of Romeo
and Juliet, you can ask: "How many movies or plays do you
know that are takeoffs on Shakespeare’s Romeo and Juliet."
Or, after defining "amplifiers" ask what political,
social, mechanical, and religious mechanisms fit the definition
Can you ask questions leading the class
to discover the concept for themselves (Socratic method)?
This works great for the students who participate.
I suspect it doesn’t do much for the students who are
waiting to be told "THE ANSWER".
Can you demonstrate the concept?
With a demonstration, with the aid of
the class, with slides, with movies, with videos, over the web...
Can students do part of the presentation?
Generally student presentations are technically
awful (not always, some easily put professional teachers to
shame), but they may more than make up with empathy what they
lack in gloss.
Can you connect the concept to an overall
theme of the lecture?
Everyone learns items best that are connected
in their mind. If the lecture has a theme to connect the parts,
students may easily remember otherwise disconnected and arbitrary
items. The theme doesn’t even have to be "correct",
just mnemonic (e.g. taken alone, "Animal Farm" appears
to be just a series of amusing and horrifying incidents. Taken
as an analogy to the effect of Communism on a country, the outrages
follow history and necessity like clockwork).
Can you connect the
lecture to the overall theme of the course?
Keep checking how the class as a whole
is doing: For example: “Has anyone got any data yet?”,
“What percent errors has anyone got?”, “Jane
found that there is less friction if you . . . “, “Did
you remember to level the equipment before taking data????”
About note taking! Don’t expect much. Strange!
Stranger yet, students generally only take notes by copying what
you put on the chalkboard, completely ignoring all the clever
hints you give vocally. My daughter suggests you write: “Take
notes of what I say.” on the board. Have
the equipment for the students to handle as you explain /
demonstrate. Ask the students to find and explain
the parts. Demonstrate some functions, then give
the students time to emulate you. Leave your setup working
for the students to examine. Don’t expect the students
to remember long, detailed explanations.
Have next week’s lab set up for the students to examine.
Have the students put their results on the chalkboard as they
progress. Then everyone can see what is reasonable, what averages
are, perhaps compete for the “best” results.
White Board: Have the students do their calculations
on a large (12” by 18” say) smooth white board. They
can write with dry erase pens intended for classroom white boards.
With these large calculations everyone in the group (and you)
can easily see what is going on.
and consulting: What do you and the students do after the lab?
Why lab reports?
No experiment is a complete waste. It can always
be used as a bad example. But, if the experiment is not
reported, or the report is unintelligible, even the most brilliant experiment
was a complete waste. If you can't stop a passerby on
the street and explain your experiment then you don't really
understand it yourself. Most students and many professionals
would prefer a root canal to sitting down to write a report
on an experiment. Many teachers feel that time spent
in report writing is time taken away from the experiment.
"The students can learn about report writing in a later
course." In contrast to some of these views I feel it is
never too early to start learning about reporting results; in
fact reporting is an integral part of doing and learning from
- First, if report writing is put off until "some
later course" the student may never be forced to learn
how to write reports at all!
- Second, it is probable that the student doesn't really
understand the experiment if they can't produce a lucid description
of what was done and why it was done that way. There
is a fair chance that in trying to explain what happened,
the student will deepen their own understanding and become
aware of aspects of the procedure or results that would otherwise
have escaped notice.
- Third, if there are questions later as to just what
did happen, it is vital that the report contain the
necessary information: equipment, dimensions, procedures,
details of the analysis,. . . . . . If anything
interesting is found there are always questions.
- Ultimately perhaps most important to any professional
is that progress in their field depends dissemination of the
test results (or what people know about your work).
There are no Nobel Prizes for unpublished work. There
is no long employment for even the most brilliant researcher
if no one in the company can understand their reports.
Length and time for lab reports: There are limits
as to how much time students can be expected to (or should,
considering that they may have a life outside the physics lab)
spend on report writing. It seems to me that most of the
time should be spent in thinking and not in the mechanics of
producing the report, so shorter thoughtful reports should be
encouraged over long, rambling, pointless reports. (e.g.
untouched computer printout.) Many students have a great
deal of trouble distinguishing between the value of polish and
volume in contrast to thoughtfulness. Considering what
sells, this confusion is quite reasonable.
Work finished in the lab:Much of
the report can be done in lab if slightly sloppy (but clear
enough) work is acceptable. Equipment lists, sketches
giving dimensions, data tables, many plots, and a preliminary
conclusion can be prepared during the lab.
Initialing parts of labs as they are done: If
you insist on initialing a rough draft of the conclusion during
lab, you can often assure that the students are headed in the
right direction and at least understand the point of the lab!
If the conclusion is at odd with reality (or the student's data),
you can encourage them to rethink what they are saying.
With the equipment still available to check again, many otherwise
hopeless results can be saved by locating faulty measurements,
bad readings, misreported data, etc. This, finally, reinforces
one of the vital lessons to be learned by doing labs:
"Trust nothing! - Check! Check!
During the week consulting: Encouraging
students to check with you during the week before handing
in labs can be very rewarding. I tell my students that
before the lab is handed in my advice is free, after the lab
is handed in my advice, if needed, will be paid for with lower
grades. Discussing labs before they are handed can result
in much lower stress (for the student) since they are just trying
to see if they have the correct idea and are not being told
they handed in something WRONG. You can communicate much
better with the students in this informal session than by exchanging
reports and written complaints on the reports. You can
see how puzzled the students look and they can easily ask for
more explanation, examples...
Content of lab reports:
The Procedure section in reports: In contrast
to scientific writing constrained by the space constraints of
journals, it is very useful to encourage students to present
their procedure in a narrative form describing the motivation
for procedures, wrong turns, cures, successes, failures, puzzles,
etc. The fact is that this narrative type of presentation
is much more useful to anyone planning to repeat the experiment
than a cold, perfect description of the final "successful"
procedure. This narrative also gives you, the teacher,
much more confidence that the students were actually there and
thinking while the experiment was done.
"Interesting" results: I emphasize
to my students that for class purposes an interesting failed
experiment is better than a dull (straightforward) experiment.
One of my more perceptive students pointed out to me that it
was actually better to get weird results because they gave an
opportunity for interesting analysis while the "correct"
results didn't leave anything to talk about in the report.
"Correct Results": Students have undoubtedly
been taught in other courses that they will be rewarded for
reporting the "correct" results (irrespective of what
actually happened). It is generally a considerable cultural
shift to convince the students that they will be rewarded for
reporting the actual results.
Of course the students with "interesting" results
must also recognize that the results are noteworthy and
try to analyze the implications. Students should be encouraged
by being told that the interesting parts of science are the
parts that we can't explain yet. Within the limits of
a school lab (or, indeed, the real world) there is no shame
in not being able to explain all interesting results. However,
unexpected results that are not even recognized by the student
as weird are a red flag to the teacher showing that the student
doesn't know what they are doing and should not be released
upon the world. For example a lever that gives out more
work than is put in should be treated with great suspicion.
What data to report?:
Students (like professional scientists) regularly take
data until they get the “correct” result, then they
stop and report only the data that gives the “correct”
result. Of course there is no point in reporting data that
resulted from flawed procedures in a professional report. (Actually
it probably would be useful to report, as I suggest below, but
journals now have little space for such niceties.)
However in a student report it is very helpful if the
students report what they tried, what went wrong, and what they
did to solve the problems. This narrative shows that the students
actually did the experiment and that they were thinking about
the results! I tell them that if they give it a good try and
produce an intelligent report, they can get a fine grade with
impossible data. Further, if I learn about common problems,
I can warn students so they can avoid these problems in the
Returning labs for a better grade.
If lab reports can be a dialog between the teacher and
student, rather than a one-way street, the students can learn
from their mistakes and get a grade commensurate with their
new knowledge. I encourage students to return improved
reports within a week after I return the graded and marked reports
to them. There is no limit on how high the final grade can
be: D- to A+ is wonderful.
To keep the work to a minimum for the students and myself,
I hand out the following rules:
returning labs for a better grade
IN DOUBT: READ THE INSTRUCTIONS
|| Labs must be returned to
the teacher the next meeting after they are returned to you.
|| There must be a significant
improvement, or attempt at improvement, on the items listed
by the teacher. No improvement, no more returns.
|| Write the date you are
returning the lab. i.e. "Returned 10/3/03"
|| Explain how to find the
changes (e.g. "in green ink", "on pages 1-3",
"circled", "on pages labeled NEW")
|| Fix what the teacher complained
about first, then make other changes if you want.
|| You generally do not have
to redo everything, just the faulty parts, reuse the good
|| Include the original report,
particularly the original list of problems.
||If you write the report
on a word processor show the changed or new parts in a different
type, e.g. italics or boldface.
||The finished report must
be arranged in the proper order; add your changes in the logical
Law: The fundamental law of experimental science: "Anything that
can go wrong will go wrong."
one: "And, at the worst possible moment."
This law is best exemplified by Murphy's
tragic death. He was visiting London where American visitors
often get in trouble because, from habit, they look for traffic
to their left when they cross a road. Murphy however remembered
to look to the right where British traffic would be approaching.
He saw nothing to his right, started across the street, and
was struck dead by an American tourist driving down the wrong
side of the road!
Some of the equipment doesn’t work (one group):
Usually, luckily, the equipment doesn’t work because
the student forgot to plug it in, forgot to turn it on, . . .
Insist that the students call you over to check the equipment
before they take it apart and replace it with other equipment.
Usually you can fix the problem with a click or two.
Check list for
debugging one piece of equipment
a power light on?
change when you turn the knobs?
knobs in a foolproof setting ( i.e. the simplest possible
knobs set the same as some similar equipment that is working?
exchange the item with a working item, do things work better?
put the suspected item in a working setup, do things still
forget, even wires go bad but look good!
doesn’t work?? Replace it with another if you have
don’t have spare equipment, you can let the students
join another group.
Check list for “None of the equipment
in the class works”
||Did you PRACTICE, PRACTICE,
||Check for universal cures.
e.g. is all the power in the room turned off?
||Perhaps everyone is following
the same bad advice – some student’s, yours, or
even the lab manual’s. Rip one of the set ups completely
apart and put it back together yourself.
||The students may not realize
how carefully they must follow the instructions to get the
experiment to work. Stop everyone and emphasize the importance
of following instructions.
||Nothing has worked yet?
Change to a lab on debugging!
||Still nothing? Change to
a lab you can do with some equipment that works.
||If the concept is really
important, change to a two week lab and try, desperately,
to get things working by the next week!
||Change the lab to “extra
credit” for the best attempts to make it work.
Check list for “Everything seems
to work, but the results are wrong (for the whole class)"
Regard this as not as a problem but as an opportunity to teach
||Point out the virtue of
checking the results as you go!
||Ask the class why
the results look wrong.
||Ask the class what could
be causing the problem.
||Do some brainstorming for
likely to wildly improbable causes. Almost anything may eventually
turn out to be the problem. Be sure to include bad data,
bad calculations, and bad theory! Often everyone is copying
an incorrect procedure or calculation! See if the class can
figure it out.
||Write calculations out on
the board with the help of the class. Beware of mixed units!
( e.g. If one dimension is much larger than
another, it is very easy to measure one in meters and the
other in mm!)
||Can you find some stage
(from initial measurements to final conclusion) where the
data looks ok? If so, you can work toward the end and usually
find where things went wrong.
Check list for wrong results for just one group:
||Congratulate them for checking.
||See if the data looks
||Examine their NEAT calculations
and sketch to see if they were going at it correctly.
||Still looks ok? Have different
members of the group make the measurements.
||Still won’t work?
Have them compare their data with the data of a group that
got reasonable results to see if there are any gross differences.
Checklist for common problems:
||Measuring the wrong thing
||The instruments were read
incorrectly (e.g. oscilloscopes are labeled in
volts/interval but the user is supposed to know that it is
the large intervals that count, not the small ones!)
||A “Sanity Check”
can eliminate some errors (i.e. having the students measure
some known quantity). For voltage measurements, a flashlight
battery has about 1.5 volts. If the measurement is 15 volts,
there is clearly some error. Varner calipers can measure
one centimeter on a meter stick, . . .
||Instruments may not be set
to give calibrated results.
||The range buttons on instruments
may not be understood. Try a sanity check.
||The students don’t
notice or don’t know the meaning of multipliers such
as mu (millionth), m (thousandth), etc.
editable Word version of the checklist.
Remember Murphy’s Corollary for debugging:
“The part that absolutely, certainly, positively
can’t go wrong, will”
Safety for the students and public:
What aspects of the experiment can harm students or the
public? I.e. laser light, falling objects, dangerous advice,
. . .
How can you minimize the possibility of harm?
- Use safer equipment (e.g. many schools are using electronic
thermometers in place of the traditional mercury in glass thermometers
to avoid the danger of broken glass and mercury exposure).
- Use the minimum voltage, current, temperature, speed,
etc. that will make the experiment work. We have several experiments
that require high voltages that the students could easily touch.
We give safety instructions of course, but the saving grace is
that the high voltage supplies produce only a very small
current. At worse the student (or teacher) gets an unpleasant
shock but does not end up dead.
- Redesign the physical set up (e.g. to keep laser light
out of the student’s eyes we keep the lasers at waist level).
- Train the students in safe practices. This takes time
Perspective: When I taught driving, many
of the students were teenagers just old enough to get their
license. They learned very quickly, in fact this was actually
a problem! The typical teenager could do everything necessary
to pass the test for their license with about ten hours of
instruction. Unfortunately this short time meant they had
not spent much time on the road and had not encountered many
of the emergencies that must be dealt with while driving.
If the student got their license at once, as seemed logical
to them, they would have to learn to deal with these emergencies
on their own with no instructor to look out for them. Within
the family we made sure the teenagers had at least fifty hours
of instruction (or at least practice with a licensed driver)
before they set out driving on their own. This seems to be
the law now in California.
In the lab, for example one using lasers, it is necessary
to not only tell the students what good safety practices are,
but to continually walk around and see that the students are
not looking into the lasers, aren’t putting their heads
at laser level, aren’t putting the lasers at eye level,
aren’t (accidentally usually) pointing the lasers out
into the hall to zap people passing by, that they do have something
in place to block the laser if it gets past their experiment,
that they close the shutter on the laser to block the light
when the laser isn’t being used temporarily, etc. You
have to continually check on these things, not just once but
every time the lasers are used and every hour or so during the
Know how to phone for help.
Be sure there is a phone available. If there is
no wired phone a cell phone, may well be worth considering.
Be sure who you will get when you dial 911. On our campus
911 gets us campus security, not the paramedics etc. that we might
Safety for the equipment:
Experience is directly proportional
to equipment ruined.
When you are worrying about risking expensive equipment
with students, it may help you to reason that one justification
for schools is that by the time students are unleashed on the
world the students will have already had lots of experience “trying”
to destroy equipment and will therefore be less likely to destroy
rare and vital equipment at their new job.
Expect that, despite the best of instruction, students
will connect and adjust equipment in any way that is physically
possible, and in ways you could have sworn were not physically
possible. For your sanity the equipment must be able
to withstand whatever may happen to it.
However, you can insist that fragile equipment (e.g. lasers)
not be balanced on shaky stacks of textbooks and lab jacks. If
equipment must be raised, show the students how to make sturdy
stands. If necessary you can buy or build the parts necessary
to keep the equipment safe. I feel that it is good training for
the students to design and build as much of the equipment as possible
within the limits of time and skill. The students will then have
a much better appreciation of the function of the equipment and
the challenges and opportunities of engineering.
Sometimes it will be necessary to physically bolt down
equipment so it won’t be brushed onto the floor (e.g. computer
monitors on carts). Electrical connections to equipment
on carts are an invitation to disaster. If equipment on a cart
is plugged into the wall, it is inevitable that someone will move
the cart without unplugging the wires and drag the equipment onto
the floor. We bolt a strip of plugs to the cart and plug it
into the wall, and then we plug the equipment on the cart into
this secured strip of plugs.
About waiting for instructions:
Whenever the class gets equipment new to them I ask “What
is the most important thing to do when you first get equipment?”
I’ll get a variety of responses, mostly reasonable such
as “plug it in”, “read the manual”,
etc. I then tell them the most important thing to do with new
Don’t touch it until you get
instructions! - The class laughs, but they
often remember too!
Equipment eventually fails on its own, so one doesn’t
want to blame students for the “natural” death of
equipment. On the other hand, students should feel responsible
for ignoring instructions and destroying equipment.
We have had good luck with the following rules:
- If the equipment is broken when the students get it,
the students are not responsible. As soon as possible after
receiving instructions, students should check that the equipment
- If the students bring back the equipment, as “it
doesn’t work” at the beginning of the class, we
explain we will figure the equipment was broken when we gave
it to the students.
- However, if the students bring back the equipment as
broken at the end of class we may well figure they broke it.
- If, as far as we can tell the students were following
instructions when the equipment died, we will assume a natural
- Finally, if the equipment dies because the students
were not following instructions, the students are responsible
for replacing the damaged equipment. Actually, we have never
had a student destroy a thousand dollar oscilloscope by dropping
it, or frying it with high voltage; perhaps our threats have
You might want to look at the debugging section here, which
gives hints on how you and the students can check whether the
equipment really is bad.
The need for the latest and greatest
If you are training students to go out to industry and
immediately start to work, it is vital to have current equipment.
It is simply misleading to students to train them on obsolete
However if the object is education, equipment is very secondary
to the design and philosophy of the course. Properly educated
students may not know exactly how the latest gismos work, but
the students do know what is possible, what is necessary,
what to ask, and how to educate themselves about the latest details.
Often in the real world one must make do with what is available.
If training must be done and good equipment isn’t available,
wonders can be done with what appears to be unsuitable equipment.
Of course you must let the students know what the situation
is: “This equipment isn’t the real stuff, but you
will learn the principles so you will know what is going on when
you get the real stuff.”
Perspective: The most spectacular example I
have been associated with was conducting a program for Laser
Technicians before we had any lasers. What we didn’t
have we simulated. Instead of doing interference experiments
with lasers, we did them with water, sound or microwaves. Instead
of commercial lens testers we made crude ones out of fiberboard
and cardboard. Surprisingly, the students were the most successful
students we ever had. Education instead of training perhaps?
Equipment overkill, or why you might not really want that
fancy equipment you could buy with that new grant:
The point to keep in mind is that, aside for training for
specific equipment, education requires that the students completely
understand what they are doing and be able to generalize it to
other situations. Often this means (as far as equipment goes)
”simpler is better”. On one end of the
spectrum is a spectrometer that students construct themselves
with meter sticks, clamps, lights, and a prism. There are few
parts, and all the parts are right out in the open to see and
manipulate. On the other extreme is a computerized
spectrometer that just requires the push of a button to calibrate
itself, take a spectrum, and plot the results. It might even
do a fast Fourier transform to process the data!
For 99.99 percent of the students, using the crude spectrometer
will lead to much greater understanding! The other 0.01 percent
of the students are at Caltech or MIT and don’t need us
anyway. Even the 0.01 percent student will probably have fun
improving the crude spectrometer!
Computers have been a mild curse in the educational equipment
market. Once a computer is available, it is overwhelmingly tempting
for the programmers to have the computer do all the amazing things
computers can do, leaving mere humans in its dust. This is great
if you are using the data for research, but isn’t so hot
if you want students to understand what is happening and why.
A Faustian bargain?
Modern equipment steadily becomes faster, more sensitive,
more rugged, and easier to use. Many effects that for generations
had been the stuff of blackboards are now easy to do experimentally
in a clear, elegant fashion. Unfortunately, in some
cases the experiments may be more clear and elegant to the teacher
than to the students. If the students only see “black boxes”
that produce incomprehensible numbers by mysterious means we haven’t
achieved much education.
On the other hand, often the sensitivity or speed of more
modern equipment permit good measurements of phenomena that previously
gave untrustworthy results (because we were working on the ragged
edge of the capabilities of the equipment). Now we have results
that are much easier for the students to understand. In the nature
of things (some corollary of Murphy’s law) the most interesting
effects are always just beyond our current capabilities.
If we are blessed with new, expensive equipment and computer
programs we may be teaching the students to depend on tools that
they and an ordinary company will not be able to afford. It is
good to always consider using programs that are widely available
and to teach with industry standard types of equipment.
Perspective: This was recently brought home
to me when I proudly sent an uncle (a retired math teacher)
a set of Microsoft Excel programs I had written (for doing non
linear least squares curve fitting) with hyperlinks to a Microsoft
Word help file. He e-mailed back that he owned neither Word
nor Excel but had been able to try them out at the local library!
Although all the computers at the school may have certain programs
not everyone in the real world does. It’s probably safest
to assume that your readers may have only a web browser. Those
with the time and talent to write programs in Java can pass
them on to everyone with a browser, the rest of us are more
constrained. I feel, strongly, that becoming familiar with industry
standard programs and equipment and using it to build up many
different experiments is immensely better for students using
for each experiment special purpose programs and equipment (which
will never be seen again) .
In the real world (outside of government boondoggles)
one makes do with what equipment is available. To paraphrase
the army saying: "A good experiment today will always beat
a perfect experiment tomorrow."
Students, given a good example, will soon learn to
improvise with the equipment available instead of whining for
newer, better equipment. It’s worth
No one will pay them BIG BUCKS for
doing easy things.
Lab manuals are a puzzle. Before we had manuals for most
of our courses, we often heard the reasonable complaint from students
that “How can we prepare when there is no lab manual?”
. After we (I) wrote manuals incorporating most of the items below,
did the students gratefully drink in the knowledge presented in
the manuals? Guess!!
Still, lab manuals can have many vital and enriching types
“Modern formats for lab manuals” CDs and Web based:
There are several attractive aspects to offering lab manuals
on the web or on CD. The content can easily include pictures,
sound, and motion. If the content is on the web it is “easy”
to update instantly, in contrast to printed manuals or CDs. Downsides
to computer-based manuals would seem to be that the “manual”
may not be available in lab, or indeed the many places students
may study. It may also be hard for students to make notes to
themselves on CDs or web sites! A very appealing
upside of web-based material could eventually be worldwide sharing
of material. It is very appealing to think of simply linking
to a site with a splendid explanation of the standard deviation
of the mean rather than laboriously trying to reinvent this (very
The following sources may provide good ideas (most of these
sources will provide rather “bare bones” manuals with
little motivation, history, tools of the trade):
- Commercial lab manuals: Generally not great but
- Free (or almost free) lab handouts from equipment manufactures:
Usually quite good for using the manufacture’s equipment.
- String and sealing wax” manuals for grammar school:
There are many books of these experiments; many are really
- Do it yourself: Good but time consuming!
- Teacher’s journals: Your teacher’s
journals may have many generations of fascinating suggestions.
The main problem is finding time to digest all the great ideas.
- Challenge the students to invent and develop the lab:
These are the best labs of all, fun for students and teachers.
Many of our labs evolved over many generations of student improvements
and inventions. The downside is that the teacher
must be constantly involved to head off dead end plans.
Scheduling of lab topics – lecture or lab first?
For an individual, useful knowledge evidently develops
in two complementary ways. On the one hand a broad overview (world
view or theory) can suggest expansions of the known facts to new
possibilities that are tested by experiments. If successful,
these new “facts” are easily incorporated into the
student’s existing worldview. If the learner’s original
worldview is incorrect, then (experiments in learning show) considerable
concrete, factual evidence must be adsorbed before the worldview
can be corrected. In contrast, if the learner has
no preconceptions (no existing worldview on this subject) concrete
observations may gradually be incorporated into a cohesive worldview.
Both procedures seem valid and necessary.
Theories of learning seem to favor the first plan (incorporate
new “facts” into existing mental structures). However
almost all experiments in learning physics seem to imply that
it is necessary to have concrete (hands on) examples in order
to build up a useful (problem solving in this case) worldview.
So, there are justifications for presenting concepts first
as theories, presumably in lecture, or presenting the concepts
first as the results of experiments that the students experience
for themselves in lab. Strangely, students seem to feel that a
topic is not real until it is presented in lecture, even if they
have the same teacher for lecture and lab.
One might think that “success” would be easy to identify,
but it is often in the eye of the beholder.
- To the student: Consider the opinions of two typical
students (stolen from Galileo’s writing) explaining why
their labs were good labs.
- During the lab: Salviati: I learned lots of good stuff
and had fun doing it. Simplicio: We got out early
and didn’t have much of a write up.
- During the semester:Salviati: I find
I am learning lots about analysis, professional behavior,
and report writing. My grades reflect what I have learned.
The teacher really cares and goes out of his/her way to help
me learn. Simplicio: I’m getting good grades without
- During following courses:Salviati: This stuff I learned
in that lab (yours) really puts me ahead of students from
other schools. Simplicio: Why aren’t I doing as well
as I did in that easy lab (yours)?
- During their career: Salviati: All those
lab and professional techniques I learned in that lab (yours)
keep paying off. Simplicio: Well, that lab (yours)
was the last easy thing I ever had.
- To the teacher: Depends, I’ll generally go
along with Salviati. It is also nice if your peers appreciate
what your labs are doing for the students.
- Administrators: Varies. Sometimes “Wow,
we even have a good reputation at Harvard” at other times
“The students never complain.”
- Parents: I got my money’s worth from that
lab, my daughter really learned a lot of useful stuff.
- The Community: We are getting our money’s
worth. Our students are encouraged, do well when they transfer
or get jobs and are taught good moral principles as part of their
- Transfer Institutions: Great, students from ------------
do better than our native students. (Don’t laugh, this
is true in many instances.)
- Professional groups (Unions, Nurses, etc.): Students
from ------------ really know their stuff, they almost always
pass our tests and have good, professional attitudes and habits.
Good bye, Good luck, Have fun!