On-Line Seminar, Fall 1997

Lesson 2: Methods of Analysis and Evaluation

Bibliographical Format for Citing References

As a scientist, you need to accurately document sources that provide background information used to develop your research project. Follow the MLA guidelines and examples shown below to cite the various types of reference materials. All of your sources should be listed alphabetically in your final bibliography. Although it is not shown in the examples given, the second line (and each subsequent line) of a cited reference MUST be indented.

1. BOOKS
   Include:
   
   • Author (last name first). the underlined book title. Place of publication: Publisher, Date of publication.
   
   
   
   • Example (one author): Lorant, Stefan. Pittsburgh: The Story of an American City . Lenox, Maine: Authors Edition, Inc., 1975.
   
   
   • Example (two or more authors): Hailey, Arthur and John Castle. Runaway Zero-eight . New York: Doubleday, 1973.


2. ENCYCLOPEDIAS
   Include:
   
   • "Title of article." the underlined name of encyclopedia. Edition.
   
   
   
   • Example: "Shark." The World Book Encyclopedia . 1986 ed.


3. MAGAZINES
   Include:
   
   • Author (last name first). "Title of article." the underlined title of magazine Date(for weekly magazine place day before month): Pages(s).
   
   
   
   • Example: McGuigan, Cathleen. "Born to sing." Newsweek 17 Feb. 1986: 72.
   
   
   • Example(without an author): "Whitney Houston." Teen July 1986: 51.


4. CD-ROMs
   Include:
   
   • [If it's a book on CD-ROM, include: Author (last name first). the underlined book title. original date of publication.] Title of CD-ROM. Place of publication: Publisher, date.
   
   
   
   • Example: Williams, T. Harry. The Military Leadership of the North and South . 1960. US History on CD-ROM. Parsippany, NJ: Bureau Development, 1990.
   
   
   • Example(without an author): The Bible. CD-ROM Dataset-The Old Testament. Parsippany, NJ: Bureau Development, 1990.


5. DATABASE SOURCES
   Include:
   
   • Author (if available). "Title of article." the underlined title of database. Provider, date, record number.(s).
   
   
   
   • Example: Nevin, John J. "Doorstep to Freetrade." Harvard Business Review 61 (1983): 88-95. DIALOG Information Services, 1983, record no. 83-N43.
   
   
   • Example: "Alexander Hamilton." Academic American Encyclopedia. 1981 ed. Compuserve, 1983, record no. 1816.


6. TELEVISION OR RADIO PROGRAMS
   Include:
   
   • Author(if any). the underlined title of program. City where broadcast from: station call letters, date of broadcast.
   
   
   
   • Example: Shakespeare, William. As You Like It . Nashville: Nashville Theater Academy, WDCN-TV, 11 Mar. 1975.
   
   
   • Example: The Commanders: Douglas MacArthur . New York: NBC-TV, 17 Mar. 1975.


7. VIDEOTAPES
   Include:
   
   • Author or commentator. "Title of the selected segment." or the underlined title of entire videotape. Name of video series. Location of publication: Publisher, date.
   
   
   
   • Example: Thompson, Paul. "W.B. Yeats." Lecture on Videotape, VHS-MSU 160. Memphis: Memphis State U, 1982.
   
   
   • Example: Sevareid, Eric. CBS News . New York: CBS-TV, 11 Mar. 1975; Media Services Videotape 1975-142. Nashville: Vanderbilt U, 1975.


8. INTERVIEWS
   Include:
   
   • Name of person interviewed (last name first). Interview with person's job title. Location and date of interview.
   
   
   
   • Example: Norfleet, Joe. Interview with the county historian. Dalton, GA 6 July 1990.


9. LETTERS
   Include:
   
   • Name of person responding(last name first). Letter to the author. Date.
   
   
   
   • Example: Wilfong, Richard, and Wynona Wilfong. Letter to the author. 14 March 1990.


10. E-MAIL
    Include:
    
    • Author of e-mail message(last name first). Subject line of the message. [Online] Available e-mail: recipient address from author address, date of message.
    
    
    
    • Example: Taylor, Barry. Hubble Space Telescope image enhancement techniques. [Online] Available e-mail: student@emmaus.high.edu from btaylor@hst.nasa.gov, January 23, 1995.


11. WORLD WIDE WEB SITES
    Include:
    
    • Author (last name first). Title of item. [Online] Available http://address/filename, date of document or download.
    
    
    
    • Example: DiStefano, Vince. Guidelines for better writing. [Online} Available http://www.usa.net/-vinced/home/better-writing.html, January 9, 1996.


12. GOPHER SITES
    Include:
    
    • Author (last name first). Title of gopher item. [Online] Available gopher: address, path, date of document or download.
    
    
    
    • Example: U.S. Department of Agriculture. Agriculture Statistics for 4th Quarter 1995. [Online] Available gopher:agri.usda.gov, Department of Agriculture/Latest Statistics for 1995/4th Quarter Folder, January 28, 1996.


13. FTP SITES
    Include:
    
    • Author (last name first). Title of item. [Online] Available ftp: address, path/filename, date of document or download.
    
    
    
    • Example: Hess, Hanna. Networking in the Information Age. [Online} Available ftp:194.335.23.19. pub/research/internet/network.txt, February 5, 1996.


14. TELNET SITES
    Include:
    
    • Author (last name first). Title of item. [Online] Available telnet: address, path, date of document or download.
    
    
    
    • Example: Brady, Larry E. Map of Iraqi Troop Movements for 1/9/96. [Online} Available telnet: fedworld.gov. Government Information/CIA/Maps/Latest Maps/Iraq, January 10, 1996.


15. ONLINE CHAT (IRC)
    Include:
    
    • Name of online speaker (last name first). [Online] Available IRC: telnet , IRC channel name, date of session.
    
    
    
    • Example: McBane, Lisa. [Online] Available IRC: telnet:world.sensemedia.net: 6677,#egypt: March 8, 1996.


16. USENET GROUPS
    Include:
    
    • Author (last name first). Title of Item. [Online] Available usenet: group, date of posting.
    
    
    
    • Example: Brown, David. Educational Insights 1995. [Online] Available usenet: k12.ed.research. December 27, 1995.

1. Gibaldi, Joseph and Walter S. Achtert. MLA Handbook for Writers of Research Papers . 2nd ed. New York: The Modern Language Association of America, 1984.


2. "Technology Connection-The Book Report-Library Talk." Classroom Connect March 1996: 9.


3. Walker, Janice R. MLA Citation Guide. [Online] Available http://www.cas.usf.edu/english/walker/mla.html, August 18, 1997.

Tutorial for Scientific Measurement and Analysis

This portion of the lesson is designed to provide some background information to help you perform accurate data measurements and experimental analyses of the results. The reliability of your conclusions depends on your ability to successfully complete these scientific skills. There are some practice exercises included to check your understanding of the concepts involved.

Quantitative vs Qualitative Observations

Physical and chemical properties of matter, and the changes that it undergoes, can be described in two different ways. Qualitative observations relate the general appearances and behaviors of the matter in descriptive terms. Characteristics, such as color, odor, texture, or phase of matter are qualitative. Quantitative observations involve measurements and calculations that give more specific information, such as mass, volume or density. All quantitative measurements are comparisons to some standard unit of measurement that always has the same value. The measurement must have BOTH a numerical value and a unit value. For scientific measurements, the metric (SI) system is preferred, because every unit in the system is 10 times the size of the next smaller unit, and the names of every unit are derived from using a prefix with the base unit name. Study the chart of the most frequently used units and prefixes below:




Changing a measured value from one metric unit to another involves simply moving the decimal point location in the number. If the units get larger in value, then the decimal point is moved to the left to make the number smaller by an equal power of 10. If the units get smaller in value, then the decimal point is moved to the right to make the number larger by an equal power of 10.
Example: 3.42 meters = 0.00342 kilometers = 342. centimeters
Problem :
Determine the numerical value for the following metric-metric conversions:

All of the basic units are interrelated in that 1 gram of water(at 4¼C) has a volume of exactly 1 cm3 or a capacity of 1 mL. (Also, 1 kg water = 1000cm3 = 1000 mL =1 liter = 1 dm3)

Principal English System-Metric System Conversion Factors:

Accuracy, Precision and Significant Figures

All measured quantities will have some degree of uncertainty due to the limitations of the measuring tools or the ability of the person to use them correctly. Anytime a measurement is recorded, all digits that are certain plus one uncertain (estimated) digit are included. These digits that help to express the value of the measurement are called significant figures. Two identical nails are placed next to different centimeter rulers below. Record the length values you can obtain from reading each of these rulers, using as many significant figures as possible.



length of nail with Ruler A = __________cm .......... length of nail with Ruler B = __________cm

Since Ruler A has smaller divisions (0.1 cm), there is less uncertainty in the measured value. The length is greater than 4.3cm, but less than 4.4cm. By estimating between the divisions, a possible length could be 4.37cm, with the 7 digit being doubtful. This measurement has 3 significant figures. Student answers could range between 4.34cm to 4.40cm, which means the uncertainty is + or - 0.03 cm. With Ruler B having larger divisions of 1 cm each, the length of the nail appears to be less than 4.5cm, but it can only be estimated to be either 4.3cm or 4.4cm. The uncertainty is at least + or - 0.1cm. There are only 2 significant figures in this value, which makes it less precise.

Precision refers to the uncertainty in a measurement. The less the uncertainty, the higher the precision and easier it is to reproduce the measured value again and again. The + 0.1cm difference when using Ruler B might seem trivial, but this would be a huge error if you were measuring a smaller distance, such as the thickness of a piece of paper.

A better reflection of uncertainty is relative uncertainty (also called relative precision or relative error). The uncertainty is divided by the total amount measured. For the measurements with the two rulers:

A piece of paper has a actual thickness of approximately 0.0160cm. Therefore, with Ruler B the

The paper thickness measurement has no reliability of being correct when using Ruler B.

The accuracy of a measurement is how close it comes to the actual or accepted value. A micrometer is used to determine the actual length of the nails to be 4.3800 with an uncertainty of + or - 0.0001cm. The accuracy of the measurement from Ruler A is calculated as a percent error which indicates how far away the measured value is from the actual (or accepted) value.



Problem :
Calculate the relative uncertainty for the micrometer measurement and the percent error for the measurement from Ruler B.

Exact numbers have an infinite number of significant figures. For example, there is no uncertainty in the number of centimeters in 1 meter. However, in any recorded observation it is necessary to know the number of significant figures. Sometimes zeros are used solely for spacing the decimal point and are not part of the actual measurement. If a recorded distance is 100 meters, you need to know if the person recording the data counted by 1, 10 or 100 m increments.

The Atlantic-Pacific rule can be used to determine which digits in a recorded value are significant.

• If the decimal point is present, start from the Pacific side of the number and move right to the first nonzero digit. This digit, and every one to its right, is significant.



• If the decimal point is absent, start from the Atlantic side of the number and move left to the first nonzero digit. This digit, and every one to its left, is significant.

The number of significant figures in a calculated value will be limited by the least precise number used in its calculation. When multiplying or dividing numbers, the answer contains the same number of significant figures as the measurement with the least number of significant figures.
Examples:

• 31.8cm(3 SF) x 24 cm(2 SF) = 763.2cm2(4 SF) rounding answer off to 2 SF it becomes 760cm2



• 8.340g(4 SF) / 4.27mL(3 SF) = 1.9532g/mL(5 SF) rounding answer off to 3SF it becomes 1.95g/mL

When adding or subtracting numbers, the answer contains the same number of decimal places as the measurement with the least number of decimal places.
Examples:
49.1g(1 decimal place) + 8.001g(3 decimal places) = 57.101g(3 decimal places) rounding to 1 decimal place = 57.1g

240m(to 10s place) - 71.3m(1 decimal place) = 168.7m(1 decimal place) rounding to the 10s place = 170m

If the number to be rounded is followed by a number greater than 5, then the digit is rounded up. If it is followed by exactly 5, then the number is rounded up only if it is an odd number. With any other possibilities, the number to be rounded is unchanged.
Examples: The following numbers are rounded to 3 significant figures:



Problems :
Solve the following problems and round your answers to the correct number of significant figures:

1. 29.01m x 75m = _____________(calculated) = _____________(rounded)




2. 510km / 5 hours = _____________(calculated) = _____________(rounded)




3. 0.2934L + 0.9L + 12.07L = _____________(calculated) = _____________(rounded)




4. 33dg - 6.81g = _____________(calculated) = _____________(rounded)




5. (7.23g - 2.8g) / 145mL = _____________(calculated) = _____________(rounded)

Graphical Analysis of Data

The purpose of most experimentation is to try to find "cause and effect" relationships between an independent variable and a dependent variable. Whether a relationship or pattern exists may not be obvious by simple examination of the data, and graphing the results will help this search for some sort of trend or consistency.
Examine the following set of experimental data comparing the mass of pennies versus the number of pennies used:



A general tendency for mass to increase when the number of pennies used increases can be easily observed, but is the change in mass proportional to the number of pennies used? Prepare a graph to answer this question, using the following steps:

1. The independent variable (number of pennies) is plotted on the horizontal (x) axis and the dependent variable (mass in g) is plotted on the vertical (y) axis.


2. Spread the data out over the entire piece of graph paper. This is done by determining the scale to be used when plotting each variable. The origin of the graph does NOT have to start at 0. Calculate the range of numbers for the x variable (largest - smallest x value), then divide by the number of blocks on the horizontal axis of the graph paper. Round this value up to the nearest unit that will produce a convenient scale. Repeat this process for the y variable.
   Example: The number of pennies (x) has a range of 4 pennies (5-1) and there are 54 blocks on the horizontal axis. 4 pennies / 54 blocks = 0.08 penny / block Rounding up to 0.1 penny / block means that every 10 blocks would represent 1 penny. Label the x axis from 0 to 5, with 10 blocks per 1 penny.
   For the y variable (mass), the range = 15.01g - 3.16g = 11.85g. Divide this value by 34 blocks on the vertical axis to get 0.35g / block. If this is rounded up to 0.4g / block , then every 5 blocks would equal 2 grams. Label the y axis.


3. Clearly label the axes and give the graph a descriptive title.


4. Carefully locate each pair of data on the graph with an x or a dot with a circle around it.


5. Draw a smooth curve or straight line to represent the general tendency of the data points.

If the points on the graph had appeared scattered, then there may be no cause and effect relationship between the variables. In this case a smooth line connect the points, indicating a cause and effect relationship. The equation form for this linear relationship is y = mx + b, where "m" is the slope of the line and "b" is the y-intercept (the value of y when x = 0). The slope (m) is calculated by selecting two data points that lie on the line, then divide the change in the y variable by the change in the x variable. The b value for this experiment = 0, because zero pennies has zero mass. The equation for the line is: y = 2.9x

Other common relationships are expressed by the graphs below:


Problem :
Take a piece of paper and fold it in quarters. Using a watch with a second hand as a timer, make as many pencil dots as you can in 1 second in the first quadrant. Count the number of dots and record the value. In the second quadrant, make as many dots as you can in 2 seconds. Record the number of dots. Repeat this process in the remaining 6 quadrants (front and back of the paper), doubling the amount of time used for each successive trial. You will have results for additional time periods of 4s, 8s, 16s, 32s, 64s and 128s. Prepare a full-page graph of your data and determine what mathematical relationship, if any, exists between the number of dots made versus time.

Statistical Analysis of Data

For some types of data the methods of graphical analysis are insufficient or not applicable, and analysis using statistics is more appropriate. Usually statistics focus on two characteristics of the data: central tendencies and dispersion (spread) of the data.

Central tendencies indicate the degree of sameness within a group of data by using one value that is a "summary" of the data in the group. The median is the middle value for a group of data that has been arranged from lowest to highest values. The mode is the value that occurs most often and the mean is simply the average value (calculated by obtaining the sum of all the values (x) in a set of data, then dividing by the number (n) of values in the set). Using all 3 central tendencies gives a better representation of the data than any single tendencies.

Sample Problem:
You want to conduct a research experiment to determine the effects of temperature on the rate of a chemical reaction. So as the control, you run 10 trials at 20¼C and time when each reaction is complete. You then run 10 trials at 40¼C as your experimental group and record the times for complete reactions to occur. The sets of data were listed in order of increasing numerical value and summarized in the chart below:


The median at 20¼C is between 24s and 28s, which is 26s. At 40¼C, it is between 21s and 22s, or 21.5s.

The mode at 20¼C is 24s which occurs four times, and at 40¼C it is 21s which occurs three times.

The mean (average) for each set is calculated by adding all the time values, then dividing by the number of trials, which is 10. The mean at 20¼C is 27s and at 40¼C it is 22s.

Just one unusally high or low value can produce a mean value that is misleading. The mean is most accurate as a representative of a set of data that is evenly or symmetrically distributed.

The amount of dispersion indicates how closely the values of a given set of data agree with each other. This spread can be shown simply as the range between the highest value and the lowest value in the set. The range at 20¼C is the difference between 34s - 22s = 12s. The range at 40¼C is 25s - 19s = 6s. The data for reaction times at 40¼C has less variation, which makes it more precise.

Another way of indicating dispersion is by calculating the standard deviation for the data set. This can be thought of as an indication of the "average" of individual data differences from the mean value of a set of data. Standard deviation is calculated by 4 easy steps:

1. Subtract each value in a data set from the mean. These differences are called deviations.


2. Square the deviations.


3. Add all of the squared deviations together, then divide this sum by 1 less than the total number of values in the data set (n-1).


4. Take the square root of this number.

For a normal distribution of data, 68% of the measurements will fall within a range equal to the mean + 1S.D.(standard deviation), 95% within the mean + 2S.D. and 99% within the mean + 3S.D. Further analysis can be done to confirm that differences in experimental results are due to the variation in conditions, and not merely caused by chance.

Click on the link to other seminars focusing on the use of laservideodisks, TI-82 calculators and CBL(calculator-based lab) units to provide additional help for using other technologies to collect and process information and data.

Presenter for this seminar:

Mark A. Case

Emmaus HS, 851 North St., Emmaus, PA 18049

The Woodrow Wilson National Fellowship Foundation webmaster@woodrow.org
CN 5281, Princeton NJ 08543-5281 Tel: (609) 452-7007 Fax: (609) 452-0066