How to teach lab report write ups

[Sebastian (a lady)]

Since I'm in the last year or two before my older kids hit high school, I'm starting to work them toward being able to write up science lab reports.

 

Does anyone have any particular suggestions for this?

 

I have found some good instructions at chemistry.about.com and at this cool science blog, but it seems like my kids need some examples of what the final product should look like.

 

I used to be good at this, and I think that I could still do a good write up, but I'm not sure that I'm getting across the main points very well.

[lisak]

My daughter uses something like the ones from Donna Young: http://donnayoung.org/apologia/general-lab-book.htm#rubric

 

I basically used these forms when we began her structured science curriculum and have changed the form as needed. It is a place to start in addition to the ones you have posted.

[regentrude]

I teach my students to keep a lab notebook. If everything is recorded properly in the lab notebook, writing a formalized report afterwards should not be hard. I would not get too hung up on the report format because it has been my experience that every college instructor will have specific, different rules for how exactly he wants the report to look. So, I feel drilling a specific procedure is unnecessary.

 

This is what I required for DD's lab notebook for physics:

- a bound notebook with all pages numbered. This is a historic requirement which was intended to prevent data falsification (because you can't remove of insert a page)

- title of experiment, date

- question explored

- theoretical background with quote (textbook, section), any equations

- a sketch (or photgraph) of the experimental set-up, labeled with the materials needed (unless explicitly stated otherwise, room temperature and atmospheric pressure are assumed)

- recording of ALL raw data

I prefer data to be recorded in the lab notebook, not just stored as files. If the data are harvested electronically, I would want a printout in the notebook

- recording of all data manipulations, calculations, averaging procedures

Any mistakes made result in crossing out the wrong calculation, but not erasing. In practice, what you thought was wrong may turn out to be your most important finding.

- an error analysis. Experimental results are meaningless unless accuracy has been determined. Sources of errors should be determined, bounds/errors for the obtained quantities need to be found.

(Easiest way to estimate errors is to replicate the experiment several times and to see how far data differ from each other.)

- graphical analysis of data where applicable. Plotting data to get calculated quantities such as slopes of curves, or behavior. Fitting to determine fit parameters. Printouts of the plots go in the lab book.

-conclusion: what did experiment demonstrate? What quantity/relationship has been determined?

 

For chemistry, things will be similar. It will be important to record conditions under which experiment has taken place quantitatively (i.e. temperatures, masses of substances; that will become part of the experiment setup)

[Sebastian (a lady)]

I teach my students to keep a lab notebook. If everything is recorded properly in the lab notebook, writing a formalized report afterwards should not be hard. I would not get too hung up on the report format because it has been my experience that every college instructor will have specific, different rules for how exactly he wants the report to look. So, I feel drilling a specific procedure is unnecessary.

 

This is what I required for DD's lab notebook for physics:

- a bound notebook with all pages numbered. This is a historic requirement which was intended to prevent data falsification (because you can't remove of insert a page)

- title of experiment, date

- question explored

- theoretical background with quote (textbook, section), any equations

- a sketch (or photgraph) of the experimental set-up, labeled with the materials needed (unless explicitly stated otherwise, room temperature and atmospheric pressure are assumed)

- recording of ALL raw data

I prefer data to be recorded in the lab notebook, not just stored as files. If the data are harvested electronically, I would want a printout in the notebook

- recording of all data manipulations, calculations, averaging procedures

Any mistakes made result in crossing out the wrong calculation, but not erasing. In practice, what you thought was wrong may turn out to be your most important finding.

- an error analysis. Experimental results are meaningless unless accuracy has been determined. Sources of errors should be determined, bounds/errors for the obtained quantities need to be found.

(Easiest way to estimate errors is to replicate the experiment several times and to see how far data differ from each other.)

- graphical analysis of data where applicable. Plotting data to get calculated quantities such as slopes of curves, or behavior. Fitting to determine fit parameters. Printouts of the plots go in the lab book.

-conclusion: what did experiment demonstrate? What quantity/relationship has been determined?

 

For chemistry, things will be similar. It will be important to record conditions under which experiment has taken place quantitatively (i.e. temperatures, masses of substances; that will become part of the experiment setup)

 

Thanks. Especially for the point that the skill of collecting and properly recording the details of the experiment are as important (or more so) than the write up at the end. Good reminder.

[karensk]

...from ds's class:

 

 

Lab Report Format (9th grade biology)

 

 

 

Throughout this semester you will be asked to write a formal lab report for a lab. This will not be for all labs but a few each 9 week period. The following is the format you will use for the formal lab report. Lab report must be typed or written neatly in black pen. Each part of the lab must be in this order. If you have any questions about the lab report, be sure to ask a day or two ahead of the due date. Although the labs are done in groups in class, the lab report is to be done by you. I want your thoughts, interpretations and analysis.

 

I. Abstract

This is a paragraph the briefly describes the experiment performed. It is the last part of the lab report that you write as it will also include the type of data collected and a simplified conclusion.

 

II. Information and Question

This paragraph will present the background information about the subject the lab is about. For instance, if the lab presents the process of osmosis, this paragraph explains what osmosis is. It ends with the question the lab is going to answer. Be specific about what this question is. This means that the lab itself needs to be read so that you know what you are doing to ask the appropriate question.

 

III. Hypothesis

This is the answer to the above question. The answer must be in the “If .... then...†format. This is very important. The statement must answer the question.

 

IV. Procedure

This is a paragraph the details the procedure. You are not to copy the step by step procedure but put the procedure in paragraph form. Keep the procedure simple, don’t worry about amounts of items, for example. Also include any safety concerns of the lab, for example, goggle use, instrument care.

 

V. Data

Include the data charts and graphs from the lab. If you are going to do these on a separate sheet, be sure that you end the previous page with the procedure and start the page after this with the questions. Graphs must be done on graph paper in pencil if not done on the computer.

 

VI. Questions

Answer the questions from the lab to help with the analysis. The answers will be in essay form. In other words, in a paragraph you will answer the questions completely, explaining your answers. Complete and informative sentences are required.

 

VII. Analysis

It should discuss the outcome of the experiment, what you discovered doing the experiment. Describe any errors that might have occurred. How might others use this information. Look for connections, relationships, patterns in your data.

 

VIII. Conclusion

It should also discuss how your hypothesis compares to the outcome of the experiment. Show how the data collected supports or refutes your hypothesis, not just whether your hypothesis is correct or not. Talk about the extreme ends of the data and how that relates to your hypothesis. Did other groups verify your data? If some data is discarded or ignored, tell why and what would you do differently.

 

In the Questions section, students are to answer questions provided by the teacher.

 

HTH!

[luvnlattes]

Here's a link to a PDF file that discusses lab notebooks and reports.

 

The PDF doesn't have a completed lab report but does give examples for each step along the way. He also has a lab grading sheet at the end. I'm thinking a student could also use it as a checklist to double check that their report has all the required elements.

[regentrude]

Just a quick comment on the "if-then" hypothesis:

some experiments really do not fit this category. It is a gross simplification that a scientist starts every experiment with a well defined hypothesis.

 

Sometimes my objective can be: measure quantity x for various parameters y and see how it depends on parameter y. Sometimes I have no clue what to expect. Sometimes nobody has thought about it before, no book or article makes predictions, I have no theory. Sometimes I have a novel material, or a sample produced under novel conditions, and I am just curious what happens if I subject it to certain procedures. The experimentalist often makes a sample, measures something, throws it to the theorists and says "please explain to me why this behaves like this - I have no clue". Sometimes the explanation comes years after the data. That is research.

I think it would be good if the student encountered some experiments where there is NO clear hypothesis before , where everything is open, where there are no preconceived notions - where he gets to "just see what happens". That does not make it a less valuable experiment than one that fits the narrow box.

[Sebastian (a lady)]

Just a quick comment on the "if-then" hypothesis:

some experiments really do not fit this category. It is a gross simplification that a scientist starts every experiment with a well defined hypothesis.

 

Sometimes my objective can be: measure quantity x for various parameters y and see how it depends on parameter y. Sometimes I have no clue what to expect. Sometimes nobody has thought about it before, no book or article makes predictions, I have no theory. Sometimes I have a novel material, or a sample produced under novel conditions, and I am just curious what happens if I subject it to certain procedures. The experimentalist often makes a sample, measures something, throws it to the theorists and says "please explain to me why this behaves like this - I have no clue". Sometimes the explanation comes years after the data. That is research.

I think it would be good if the student encountered some experiments where there is NO clear hypothesis before , where everything is open, where there are no preconceived notions - where he gets to "just see what happens". That does not make it a less valuable experiment than one that fits the narrow box.

 

Thanks. I found one lab description that included both labs to test a hypothesis and labs to demonstrate what happens under a certain set of circumstances. This latter is probably more what we're doing this year. Honestly, I'm not sure my kids know enough scientific theory to form and test a hypothesis. But they loved adding different amounts of salt to ice to make ice cream and timing which mixture froze first. And there were several questions that arose as part of this lab that could have been formed into a working hypothesis.

 

I have found a lot of places that describe the parts of the lab report, but fewer that give examples of what is sufficient to fill sections like procedure, results and conclusion. So for them, I think it's a little bit like trying to write a paragraph based only on a description of a paragraph but without a sample of some well and poorly written paragraphs to look at. I may need to do a lab write up or two of my own to show them what I think they need to be doing.

 

ETA: I don't have a science degree, but I did take a good load of basic science courses in high school and college. Most of the labs that I encountered as a non-majors student were more demonstrations than actual hypothesis testing. I can't think of any instances where I was really designing my own experiment to fit my own question and theory about something. That may be the heart of science research, but I'm not sure how present it is within the entry levels of science education. (Caveat, this is based only on my personal experience as an English major in a science and engineering core curriculum.)

Edited September 27, 2011 by Sebastian (a lady)

[Create Your Ritual]

Wonderful ideas!

[regentrude]

I can't think of any instances where I was really designing my own experiment to fit my own question and theory about something. That may be the heart of science research, but I'm not sure how present it is within the entry levels of science education.

 

Here is a very simple one for upper elementary/ middle grades:

(One of the nicest experiments DD ever did, for science fair in 4th grade). She tied a little thing to a string, tied the other end of the string to a hook in the ceiling and let it swing as a pendulum. She wanted to find out whether the time it takes for a swing (the period) depended on the mass of the thing (she used a little spice container filled with different numbers of marbles), the length of the string, and how far she pulled it to the side before letting go (the amplitude). She made an initial guess which turned out to be not fully correct.

She did this experiment with basic household supplies, measured several times, calculated averages, found how much the results differed between repetitions to estimate an error. She did three sets of experiments, in each set varying only one of her variables and keeping the others fixed.

She plotted the results, period vs number of marbles, period vs length of the string and period vs amplitude on grid paper.

 

She learned a LOT. And the experiment is simple, does not require any specialized equipment, and you can spend as much time on data analysis and plotting as you want.

[Sebastian (a lady)]

That's a great experiment. It would fit well with our history study (we just did Galileo) and is a lot less detrimental to my diet than pursuing all the variables in making ice cream.

[swimmermom3]

Just a quick comment on the "if-then" hypothesis:

some experiments really do not fit this category. It is a gross simplification that a scientist starts every experiment with a well defined hypothesis.

 

Sometimes my objective can be: measure quantity x for various parameters y and see how it depends on parameter y. Sometimes I have no clue what to expect. Sometimes nobody has thought about it before, no book or article makes predictions, I have no theory. Sometimes I have a novel material, or a sample produced under novel conditions, and I am just curious what happens if I subject it to certain procedures. The experimentalist often makes a sample, measures something, throws it to the theorists and says "please explain to me why this behaves like this - I have no clue". Sometimes the explanation comes years after the data. That is research.

I think it would be good if the student encountered some experiments where there is NO clear hypothesis before , where everything is open, where there are no preconceived notions - where he gets to "just see what happens". That does not make it a less valuable experiment than one that fits the narrow box.

 

Thank you for adding this note. Just this morning, my ds planted 3 different sets of seeds with the object of the lab being "to apply the scientific method in testing which seeds will sprout first, mustard plant, sensitive plant, or cacti." He did not think this fit an "if-then" hypothesis and had a difficult time trying to torque his prediction into that format.

[In2why]

Just a quick comment on the "if-then" hypothesis:

some experiments really do not fit this category. It is a gross simplification that a scientist starts every experiment with a well defined hypothesis.

 

Sometimes my objective can be: measure quantity x for various parameters y and see how it depends on parameter y. Sometimes I have no clue what to expect. Sometimes nobody has thought about it before, no book or article makes predictions, I have no theory. Sometimes I have a novel material, or a sample produced under novel conditions, and I am just curious what happens if I subject it to certain procedures. The experimentalist often makes a sample, measures something, throws it to the theorists and says "please explain to me why this behaves like this - I have no clue". Sometimes the explanation comes years after the data. That is research.

I think it would be good if the student encountered some experiments where there is NO clear hypothesis before , where everything is open, where there are no preconceived notions - where he gets to "just see what happens". That does not make it a less valuable experiment than one that fits the narrow box.

 

 

Excellent point! This thread is on time because I am revamping our science to become more about what we want to know and how to find out instead of an actual science study. The lab notebook is vital and I didn't even know that was what I wanted until I saw it mentioned here.

[Alessandra]

I teach my students to keep a lab notebook. If everything is recorded properly in the lab notebook, writing a formalized report afterwards should not be hard. I would not get too hung up on the report format because it has been my experience that every college instructor will have specific, different rules for how exactly he wants the report to look. So, I feel drilling a specific procedure is unnecessary.

 

This is what I required for DD's lab notebook for physics:

- a bound notebook with all pages numbered. This is a historic requirement which was intended to prevent data falsification (because you can't remove of insert a page)

- title of experiment, date

- question explored

- theoretical background with quote (textbook, section), any equations

- a sketch (or photgraph) of the experimental set-up, labeled with the materials needed (unless explicitly stated otherwise, room temperature and atmospheric pressure are assumed)

- recording of ALL raw data

I prefer data to be recorded in the lab notebook, not just stored as files. If the data are harvested electronically, I would want a printout in the notebook

- recording of all data manipulations, calculations, averaging procedures

Any mistakes made result in crossing out the wrong calculation, but not erasing. In practice, what you thought was wrong may turn out to be your most important finding.

- an error analysis. Experimental results are meaningless unless accuracy has been determined. Sources of errors should be determined, bounds/errors for the obtained quantities need to be found.

(Easiest way to estimate errors is to replicate the experiment several times and to see how far data differ from each other.)

- graphical analysis of data where applicable. Plotting data to get calculated quantities such as slopes of curves, or behavior. Fitting to determine fit parameters. Printouts of the plots go in the lab book.

-conclusion: what did experiment demonstrate? What quantity/relationship has been determined?

 

For chemistry, things will be similar. It will be important to record conditions under which experiment has taken place quantitatively (i.e. temperatures, masses of substances; that will become part of the experiment setup)

 

Exactly.

 

The Illustrated Guide to Home Chemistry Experiments, Thompson has a nice description of what should be in a lab notebook (similar to Regentrude's).

[Sebastian (a lady)]

Exactly.

 

The Illustrated Guide to Home Chemistry Experiments, Thompson has a nice description of what should be in a lab notebook (similar to Regentrude's).

 

I'll have to get that out of the library again. I was too busy to give it full attention when I had it out last time.

 

We were working on the pendulum experiment this morning. Interesting how many things we had to consider. Timing error (We had two timers, but we still have to consider how to deal with variances. Not sure if we'll just average or throw out high and low and then average or what.) As we were adding marbles, one of the kids asked if the marbles were shifting in the jar. (Good question. Maybe we need tissue as a buffer.)

 

On setup, I told them to describe what we'd put together in such a way that their grandmother in another state could reconstruct it. But as we went along, we were realizing other things that we needed to note for consistency. Like who was timing and how far our pull distance was. And how much to right down from our data or how to set up a data table. Good practice, but a lot of lessons learned by making mistakes.

 

Despite what I'd consider good high school science and a strong non-majors science core in college, I don't remember having to really think about these issues except for the one year I did a science fair. And even then I'm not sure that the process of investigation was overtly taught. Sort of fun.

[regentrude]

Timing error (We had two timers, but we still have to consider how to deal with variances. Not sure if we'll just average or throw out high and low and then average or what.)

 

several thoughts on this:

1. The major source of error will be the person's reaction time to press the timer. If you time only one swing, all this error goes into ONE period. But if you count out ten swings and measure the time for ten swings and divide by ten, you have greatly reduced the error of the individual period because the measurement error divides over ten periods.

 

2. To analyze, you repeat the above procedure many times and find an average. Then you see how far the highest and lowest value deviate form the average.

So, if your average is 1.5 seconds, and the smallest value is 1.35 sec and the largest value is 1.65 sec, you can give the period as 1.5+/10.15 sec and have thus derived a good estimate for the accuracy of your measurement.

You can show this graphically by making a dot on your graph and having an error bar that extends to the sides of the dot, according to how much error you estimated.

 

Your time measurement will become more accurate if the times are not too short. If your times are very small, try a longer string.

[Sebastian (a lady)]

several thoughts on this:

1. The major source of error will be the person's reaction time to press the timer. If you time only one swing, all this error goes into ONE period. But if you count out ten swings and measure the time for ten swings and divide by ten, you have greatly reduced the error of the individual period because the measurement error divides over ten periods.

 

2. To analyze, you repeat the above procedure many times and find an average. Then you see how far the highest and lowest value deviate form the average.

So, if your average is 1.5 seconds, and the smallest value is 1.35 sec and the largest value is 1.65 sec, you can give the period as 1.5+/10.15 sec and have thus derived a good estimate for the accuracy of your measurement.

You can show this graphically by making a dot on your graph and having an error bar that extends to the sides of the dot, according to how much error you estimated.

 

Your time measurement will become more accurate if the times are not too short. If your times are very small, try a longer string.

 

I think we will probably have to go back and do what you've suggested with measuring a multi-swing period and figuring out the average for one swing. We have a pretty big range. We had talked about doing multiple swings, but then didn't. But I think we're going to need to go back and reconsider.

 

So if this were "real" research, would you just make a notation in the lab book about changing the proceedures of the experiment and why this decision was made? Keep the first set of data in the book along with the data from the improved proceedures?

[regentrude]

So if this were "real" research, would you just make a notation in the lab book about changing the proceedures of the experiment and why this decision was made? Keep the first set of data in the book along with the data from the improved proceedures?

 

Yes, absolutely. Happens all the time: you find that your first try of a procedure did not work as well and you come up with something new. You make a note, do not erase any old data (they might turn out to be the gold mine after all), and start again with the new procedure.

Very often things do not work out on the first try, and you won't notice until you are having a good look at the data.