ADVANCED PHYSICS LAB GUIDE 2007-2008

See the sample lab for application of the following procedures.

Information on error analysis

Before turning in your lab journal for the first time, make sure you've done the following:

• On the front cover, print in large letters your name and the course name.
• Reserve the first 4 pages (front and back of first 2 sheets) for the Table of Contents.
• Number the front of every sheet with the appropriate odd number in the upper right corner.

When recording in the lab journal:

• Write only in blue or black ink and only in your lab book (not on stray papers).
• If something is incorrect or needs to be remeasured, cross it out with an X and include an explanation. Erasing, whiting-out or other obliteration of data are not acceptable scientific practices.
• It is unethical to report fictitious data or another person's data that you did not personally participate in acquiring.
• Identify all numbers with a conventional or defined symbol and/or a defining phrase.
• Include the appropriate SI units with any number.

Before beginning a new experiment:

• Start the record for any new experiment on an odd-numbered page.
• Start the experiment record with a succinct title followed by a statement of the goals of the experiment. Label this statement: GOALS.
• Record in the TABLE OF CONTENTS: the experiment title, the beginning page number of the experiment report, and the date the experiment will begin.
• At the very top of the first page of the lab, write the name of your partner. If you don't have a partner, write "No Partner''.
• If you are working on the experiment as part of a group, participate fully in all phases of the experiment with your partners. Have a healthy skepticism and be willing to speak up if you think you have a better idea than someone else. Each partner independently keeps a record of all information about the experiment.

Before designing a new experiment:

• Begin by defining (in words and with a symbol) the relevant variables in the experiment. Use easy-to-remember notation. If you use non-conventional notation and/or if you define many different variables, make a table of variables and their definitions. Label this section: VARIABLES   
• State the variable(s) you will vary in a controlled way (the independent variable), the variable(s) whose response to the independent variable you will measure (the dependent variable) and the variables you will hold constant.
• Make an educated prediction (“hypothesis”), based on your personal experience and on the results of others’ experiments of which you are aware, of what the results of your experiment will be. If you have no idea what the results will be, you should find out more about the subject of your experiment before proceeding. Search for and read literature to find out what is already known about your topic. Seek guidance from knowledgeable colleagues. Include in your report a discussion of your prediction and the basis for it, along with a list of any important references that you use. Convince the reader that you understand the subject you are investigating. Label this section: HYPOTHESIS           
• State any formulae that you will assume to be true and that you will use to relate the quantities you will directly measure to other quantities that you want to know. Also state any assumptions implicit in your formulae. If your formulae are not likely to be familiar to someone reading your report, show how the formulae are derived from familiar principles, or else provide a reference where a reader can find the derivations. Convince the reader that the formulae you want to use are appropriate for the experiment you plan to do. Label this section: THEORY

When designing the experiment:

• Start with a list of equipment you will use. Identify any apparatus numbers or letters used. These are needed in case measurements must be rechecked later. If important pieces of apparatus are not numbered, provide unambiguous descriptions. Label this section: EQUIPMENT
• Draw a diagram of the experimental setup. Draw additional diagrams, as needed, to help show how measurements will be made.
• Describe the procedure you will use to perform the experiment. Include sufficient detail so that someone reading your report could repeat the experiment without having to ask you for additional information. Discuss how you will guard against sources of systematic uncertainty to the extent that it is possible. Convince the reader that you have taken necessary care with experimental design and procedure. Label this section: METHOD
• For each quantity that you will measure, identify the likely and unavoidable sources of error in your experiment. Indicate whether these errors are likely to be systematic or random (and why you think so). For each source of error, provide a quantitative estimate (with justification) of the uncertainty (absolute or relative, as appropriate) that you expect. Your expected uncertainties might be affected by things such as the precision and accuracy of your measurement apparatus, how many times you will repeat measurement, or by your experimental setup. Convince the reader that you understand the likely sources of uncertainty in the experiment. Label this section: EXPECTED UNCERTAINTIES
• State the level of uncertainty (absolute or relative, as appropriate) that you want to achieve for the variables you will actually analyze (i.e. the dependent and independent variables) in your experiment. Using the techniques of error analysis and your estimated uncertainties, combine the different estimated uncertainties to find the overall estimated uncertainties in your dependent and independent variables. Convince the reader that you are likely to achieve your desired goal, i.e. that your overall estimated uncertainties will be less than the level you want to achieve. 

When carrying out the experiment:

• Begin your record of the actual experiment on a new page of your journal. Label this page with the heading: DATA and with the date, time and location of your experimenting.
• Record any deviations from the expected procedure that occur in the course of the experiment.
• Record all data directly into the experiment journal.
• It is unethical to report fictitious data or another person's data that you did not personally participate in acquiring.
• Identify all data with a conventional or defined symbol and/or a defining phrase.
• Use tables when possible to organize your data. Any column in a table must be headed with the name and units of the quantity recorded in the column. Only write units in the header; don't write a unit beside each number.

 When calculations and graphs are needed (almost always):

• Begin your analysis on a new page of your journal. Label this page with the heading: ANALYSIS.
• Do the calculations independently of your partners (unless stated otherwise). That means you need to have your own TI for every lab period. Compare results of calculations afterwards.
• While you do not need to show all calculations, you must show complete sample calculations so that the reader can reproduce any numbers you use beyond your original data. Label this section: SAMPLE CALCULATIONS    
• Do a complete error analysis, in which you use your actual data to estimate the uncertainties in your dependent and independent variables and in any results you derive from those variables. Include at least one complete sample calculation so that the reader can see how you did the error analysis. Label this section: ERROR ANALYSIS
• If the analysis of your data involves a graph and fit, include the results in one or more of the following ways: i) a sketch, a full page in size, ii) a printout that is smaller than the lab page size, is taped in securely (no staples!), includes a printed data table, and shows the fit results on the graph.
• Graphs include:

·         a title, usually in the form: "y-variable" vs. "x-variable". Include a phrase that distinguishes the graph from any other similar graphs in your report. Of course, you will never use the names "x" and "y" in describing specific variables. Always name the variables so as to unambiguously identify the quantity being described.

·         the name of the variable on each axis with units in parentheses after each name,

·         all the data points,

·         a line showing the fit to the data (if a fit is carried out),

·         the equation of the fit (again, no x's and y's), including the fit coefficients, with units, and rounded to the proper number of significant figures.

·         error bars on each point,  

·         a second graph (half page OK), showing the residuals of the fit. Label the graph with axes and title.

After performing the experiment:

• Complete the analysis and report as soon as possible, while the subject and procedures are fresh in your mind.  
• Write a DISCUSSION OF RESULTS, including these items, some of which will be quantitative:     

·         what you did (a brief, one or two sentence summary of this is enough),

·         what you found out,

·         what you think your results mean,

·         whether you met your goals,

·         identification of the likely and unavoidable sources of error that could have influenced your results,

·         estimates of the magnitudes of the errors in your final results

·         description of possible improvements in methods and/or apparatus that could lessen errors.

Don't underestimate the importance of the DISCUSSION OF RESULTS. This is your opportunity to convince the reader that you did a good experiment and to explain why you might not have been able to meet your goals.