Evaluations
In evaluations, you are thinking about how good your data is: how much can you trust the conclusion or the trends in your results? How good was the equipment? Could you have performed the experiment better?
You need to talk about the following four things:
Precision: how close you instrument can measure to the actual value, for example an ammeter might measure to the nearest 0.01A.
Accuracy: given the errors in your experiment, how close your data is to the real values. (Precise Instruments will give accurate data)
Reliability: how must you can trust the result? Were the repeats similar? Did you remove anomalous results from your average? Was your answer similar to the other groups?
Validity: how well did you experiment find out what it was meant to find out?
1) First off think about all the errors that could have occurred in your experiment: could you change the method or device to improve them?
Could you change the precision of your instruments? Were there any anomalies and how could they have occurred?
2) Then think about the reliability: were all your results close to the average? Did you take enough results for a reliable average? Did you remove your anomalous results from your averages?
3) Finally, think about the validity: did you find a pattern between the variables you were investigating? Would you have to do any further work to find it? Could you improve your model?
You need to talk about the following four things:
Precision: how close you instrument can measure to the actual value, for example an ammeter might measure to the nearest 0.01A.
Accuracy: given the errors in your experiment, how close your data is to the real values. (Precise Instruments will give accurate data)
Reliability: how must you can trust the result? Were the repeats similar? Did you remove anomalous results from your average? Was your answer similar to the other groups?
Validity: how well did you experiment find out what it was meant to find out?
1) First off think about all the errors that could have occurred in your experiment: could you change the method or device to improve them?
Could you change the precision of your instruments? Were there any anomalies and how could they have occurred?
2) Then think about the reliability: were all your results close to the average? Did you take enough results for a reliable average? Did you remove your anomalous results from your averages?
3) Finally, think about the validity: did you find a pattern between the variables you were investigating? Would you have to do any further work to find it? Could you improve your model?
Dataloggers are computers that you attach to a device to take more measurements. This makes your results more reliable.
Can you think of a way of making your instruments more precise? Common problems are parallax error (reading the instrument at the wrong angle) and reaction time? An example evaluation is shown below. |
Example: Evaluation of the Light bulb Efficiency Experiment
The accuracy of my temperature reading was ±1°C. There were several sources of error in the reading of the temperature: one was a parallax error in reading the thermometer itself which could be improved by using a digital thermometer or having a magnifying glass over the thermometer to get a more accurate reading. Another was that it was hard to keep the thermometer from touching the bulb which would increase the temperature reading, making it less accurate. The temperature of the water was also different throughout the water: it was hotter nearer the bulb and cooler further away. We could improve the experiment by stirring the water to make the temperature constant throughout.
The accuracy of the voltage and current readings were to ±0.01V and ±0.01A. However, the voltage did fluctuate by ±1V throughout the experiment. We could improve the experiment by taking down the voltage and current readings every 30s as well as the temperature.
The accuracy of the time measurement was ±0.5s due to reaction time when reading the stopwatch.
Our results were quite reliable because they were similar to other groups’ results. However, we could improve our reliability by repeating the experiment and taking an average.
The experiment could also be improved by using a data logger for the temperature as it would remove human error in the reading of both the thermometer and the stopwatch, improving the accuracy and allow for more frequent data to be taken, increasing the reliability.
Our experimental method was valid as it produced data that we were able to use to calculate the efficiency of our light bulb. We could improve the validity by improving the reliability and accuracy with the suggestions above.
The accuracy of my temperature reading was ±1°C. There were several sources of error in the reading of the temperature: one was a parallax error in reading the thermometer itself which could be improved by using a digital thermometer or having a magnifying glass over the thermometer to get a more accurate reading. Another was that it was hard to keep the thermometer from touching the bulb which would increase the temperature reading, making it less accurate. The temperature of the water was also different throughout the water: it was hotter nearer the bulb and cooler further away. We could improve the experiment by stirring the water to make the temperature constant throughout.
The accuracy of the voltage and current readings were to ±0.01V and ±0.01A. However, the voltage did fluctuate by ±1V throughout the experiment. We could improve the experiment by taking down the voltage and current readings every 30s as well as the temperature.
The accuracy of the time measurement was ±0.5s due to reaction time when reading the stopwatch.
Our results were quite reliable because they were similar to other groups’ results. However, we could improve our reliability by repeating the experiment and taking an average.
The experiment could also be improved by using a data logger for the temperature as it would remove human error in the reading of both the thermometer and the stopwatch, improving the accuracy and allow for more frequent data to be taken, increasing the reliability.
Our experimental method was valid as it produced data that we were able to use to calculate the efficiency of our light bulb. We could improve the validity by improving the reliability and accuracy with the suggestions above.