Errors and Uncertainties in Chemistry Internal Assessment

Size: px

Start display at page:

Download "Errors and Uncertainties in Chemistry Internal Assessment"

Delphia Griffith
5 years ago
Views:

1 Errors and Uncertainties in Chemistry Internal Assessment The treatment of errors and uncertainties is relevant in the internal assessment criteria of: data collection, aspect 1 (collecting and recording raw data) data processing and presentation, aspect 2 (presenting processed data) conclusion and evaluation, aspects 1 and 2 (drawing conclusions/evaluating procedure(s) and results). When the internal assessment criteria are applied to investigations, the issues of errors and uncertainties must be interpreted according to the level of the candidate. There are different expectations of candidates for the same criteria at standard level and higher level. Expectations at Standard Level and Higher Level Standard level candidates are not expected to process uncertainties in calculations. However, they can make statements about the minimum uncertainty, based on the least significant figure in a measurement, and can also make statements about the manufacturer's claim of accuracy. They can estimate uncertainties in compound measurements, and can make educated guesses about uncertainties in the method of measurement. If uncertainties are small enough to be ignored, the candidate should note this fact. Higher level candidates should be able to express uncertainties as fractions,, and as percentages,. They should also be able to propagate uncertainties through a calculation. Note: Standard level and higher level candidates are not expected to construct uncertainty bars on their graphs. Interpreting the Relevant Assessment Criteria Data Collection (Aspect 1) 1. Volumetric analysis acid base titration Sample extracts of typical student work are shown below. a. Determination of the percentage of ethanoic acid in vinegar Volume of sodium hydroxide for titration 1: 20.0 ml Volume of sodium hydroxide for titration 2: 40.5 ml No raw data has been collected or presented (just the calculated volumes). The level of achievement for aspect 1 of data collection is not at all. b. Volume of NaOH = = 20.0 Volume of NaOH = = 40.5 Some appropriate raw data is recorded, but other information is omitted, such as colour at end point. No units or uncertainties are indicated. The level of achievement for aspect 1 of Page 1 of 7

2 data collection is partial. c. Titration of standard HCl against NaOH and NaOH against vinegar Final volume/cm Initial volume/cm Volume of base required/cm Some appropriate raw data is recorded, but other items, such as colour at end point, are missing. No uncertainties are recorded, data is recorded to different numbers of decimal places, data is not recorded accurately (readings are only to 0.1 cm 3 ), and only one set of data is collected. The level of achievement for aspect 1 of data collection is partial. d. Titration of 5.00 cm 3 vinegar against the standardized NaOH Trial 1 Trial 2 Trial 3 Final volume/cm 3 (± 0.05 cm 3 ) Initial volume/cm 3 (± 0.05 cm 3 ) Volume of base required/cm 3 (± 0.1 cm 3 )* Colours of solutions: acid, base and phenolphthalein indicator were all colourless. At end point, the rough trial was dark pink. The other two trials were only slightly pink at end point, and remained slightly pink for at least 30 seconds. The candidate records appropriate qualitative and quantitative raw data, including units and uncertainties. The level of achievement for aspect 1 of data collection is complete. *The candidate has calculated the error in the volume by adding the two errors in the initial and final volumes. Propagation of uncertainties is only required at higher level. Standard level candidates could still be awarded a "complete" if they had not done this. 2. Gas law investigation The following examples of data collection (see Tables 1 3 below) are from a gas law experiment. Table 1 DC (aspect 1) = complete Temperature T/ C ± 0.2 C Height of column h/mm ± 0.5 mm Table 2 DC (aspect 1) = partial Temperature (T) Height of column (h) Table 3 DC (aspect 1) = partial Temperature Height of column Page 2 of 7

3 In Table 1, the candidate designed the data table and correctly recorded the raw data, including units and uncertainties. The level of achievement for aspect 1 of data collection is complete. In Table 2, units and uncertainties are not included. The level of achievement for aspect 1 of data collection is partial. In Table 3, units and uncertainties are not included, and the data is recorded in an inconsistent manner. Significant digits are not appreciated. The level of achievement for aspect 1 of data collection is partial. Note: If candidates graph the raw data directly, and do not make a permanent record of it (for example, a computer program logs and graphs the data), then they fail to meet any of the requirements for aspect 1 of data collection. The level of achievement for aspect 1 of data collection would be "not at all" as candidates should produce a printout of the raw data. Uncertainties in raw data When numerical data is collected, values cannot be determined exactly, regardless of the nature of the scale or the instrument. If the mass of an object is determined with a digital balance reading to 0.1 g, the actual value lies in a range above and below the reading. This range is the uncertainty of the measurement. If the same object is measured on a balance reading to g, the uncertainty is reduced, but it can never be completely eliminated. In data collection, estimated uncertainties should be indicated for all measurements. Data Processing and Presentation (Aspect 2) Figure 1 DPP (aspect 2) = complete Figure 2 DPP (aspect 2) = complete Figure 1 is a graph of the gas law data from Table 1, where the has a significant uncertainty. The computer drew the uncertainty bars based on the candidate entering the correct information, which in this case was 0.5 mm for each value. Figure 2 does not show the uncertainty bars. In chemistry, candidates are not expected to construct uncertainty bars. In Page 3 of 7

4 both graphs the title is given (although it should be more explicit), and the candidate has labelled the axis and included units. The level of achievement for aspect 2 of data processing and presentation for both graphs is complete. Figure 3 DPP (aspect 2) = partial In Figure 3, the candidate does not include a title for the graph, and the units are missing. The level of achievement for aspect 2 of data processing and presentation is partial. Figure 4 DPP (aspect 2) = not at all In both examples shown in Figure 4, the candidates fail to draw a best straight-line graph (or draw no line). The units and the titles of the graphs are missing. In the second graph, poor use is made of the x-axis scale. The level of achievement for aspect 2 of data processing and presentation for both graphs is not at all. Conclusion and Evaluation (Aspects 1 and 2) Page 4 of 7

When attempting to measure an already known and accepted value of a physical quantity, such as the charge of an electron, the melting point of a substance, or the value of the ideal gas constant,

Perhaps a candidate measured the value of R, the ideal gas constant = 8.11 kpa dm 3 mol -1 K -1, and the accepted value is 8.314 kpa dm 3 mol -1 K -1.

5 When attempting to measure an already known and accepted value of a physical quantity, such as the charge of an electron, the melting point of a substance, or the value of the ideal gas constant, candidates can make two types of comments: 1. The error in the measurement can be expressed by comparing the experimental value with the textbook or literature value. Perhaps a candidate measured the value of R, the ideal gas constant = 8.11 kpa dm 3 mol -1 K -1, and the accepted value is kpa dm 3 mol -1 K -1. The error (a measure of accuracy, not precision) is 2.45% off the accepted value. This sounds good, but if, in fact, the experimental uncertainty is only 2%, random errors alone cannot explain the difference, and some systematic error(s) must be present. 2. The experimental results fail to meet the accepted value (a more relevant comment). The experimental range does not include the accepted value. The experimental value has an uncertainty of only 2%. A critical candidate would appreciate that they must have missed something here. There must be more uncertainty and/or errors than acknowledged. In addition to the above two types of comment, candidates may also comment on errors in the assumptions of the theory being tested, and errors in the method and equipment being used. Two typical examples of student work are given below. Melting point determination Conclusion and Evaluation: Intermolecular bonds are being broken and formed which consumes energy. There is a definite correlation between the melting point and the freezing point of a substance. If good data is collected, the melting point should be the same as the freezing point. A substance should melt, go from solid to liquid, at the same temperature that it freezes, goes from liquid Page 5 of 7

to solid. Our experiment proved this is true because, while freezing, the freezing point was found to be 55 C, and when melting, the melting point was also found to be 55 C (see graph).

6 to solid. Our experiment proved this is true because, while freezing, the freezing point was found to be 55 C, and when melting, the melting point was also found to be 55 C (see graph). The candidate states a conclusion that has some validity. No comparison is made to literature value. There is no evaluation of the procedure and results. The level of achievement for aspect 1 of conclusion and evaluation is partial. The level of achievement for aspect 2 is not at all. Melting/freezing point of para-dichlorobenzene Melting point = boiling point = 55.0 ± 0.5 C Conclusion and evaluation: Literature value of melting point of para-dichlorobenzene = 53.1 C (Handbook of Chemistry and Physics). The fact that % difference > % uncertainty means random errors alone cannot explain the difference and some systematic error(s) must be present. Melting point (or freezing point) is the temperature at which the solid and the liquid are in equilibrium with each other:. This is the temperature at which there is no change in kinetic energy (no change in temperature), but a change in potential energy. The value suggests a small degree of systematic error in comparison with the literature value as random errors alone are unable to explain the percentage difference. Evaluation of procedure and modifications: i. Duplicate readings were not taken. Other groups of students had % uncertainty > % difference, ie in their case random errors could explain the % difference, so repeating the investigation is important. Page 6 of 7

7 ii. The thermometer how accurate was it? It should have been calibration. In order to eliminate any systematic errors due to the use of a particular thermometer, calibration against the boiling point of water (at 1 atmosphere) or better still against a solid of known melting point (close to the sample) should be done. iii. The sample in the test tube was not as large as in other groups. Thus the temperature rises/falls were much faster than for other groups. A greater quantity of solid, plus use of a more accurate thermometer (not 0.5 C divisions, but the longer one used by some groups), would have provided more accurate results. The level of achievement for aspects 1 and 2 of conclusion and evaluation is complete. back to top back to contents Page 7 of 7

11.1 Uncertainty and error in measurement (1 Hour) 11.2 Uncertainties in calculated results (0.5 Hour) 11.3 Graphical techniques (0.

11.1 Uncertainty and error in measurement (1 Hour) 11.2 Uncertainties in calculated results (0.5 Hour) 11.3 Graphical techniques (0. Chapter 11 Measurement and Data Processing Page 1 Students are to read and complete any part that requires answers and will submit this assignment on the first day of class. You may use internet sources