Plan out Experiment, identify all dangers, whose at risk, how to reduce risks and consider ethical issues.
1 of 14
Anamalous results.
Investigate, find out what happened and dismiss that result. No valid conclusions can be drawn from these type of results.
2 of 14
How to devise investigation.
Consider (in)/dependent and control variables, Negative control/placebo, ensure large enough sample, apparatus/technique, measurements/recordings and safety/ethics.
3 of 14
Standard Deviation.
Indicates variation of values in a single sample. (9+/- 3) = Mean is 9, standard deviation is 3. Fatter graph = larger standard deviation = greater variation in values.
4 of 14
Error Bars.
Represent standard deviation of charts/graphs for mean values. Longer bar = larger standard deviation = larger spread from mean value.
5 of 14
Cm^3 + dm^3.
1000cm^3 = 1dm^3.
6 of 14
Nanometres = micrometres = millimetres.
1000nm = 1um = 0.001mm.
7 of 14
Student's T-test.
Used to calculate if there is a significant difference between two sets of data. Calculated value compared to critical value, value obtained greater than critical value at probability of 55 or less, there is a 95% confidence of a significant differen
8 of 14
Chi-squared test.
Used for categorical data for statistical comparison of observed and expected results. If result larger or equal to critical value, 95% certainty of significant difference.
9 of 14
Correlation coefficient.
Used to work out degree of correlation between two sets of continuous data (ranging values from -1 to +1).
10 of 14
Drawing Conclusions From Data.
Conclusions need to be valid by answering the original question using repeatable and reproducible results. Look at correlation between two variables (bare in mind, correlation does always mean a causation between the two factors). Control variables.
11 of 14
Evaluating Methods and Results (repeatability).
Repeat experiment as many times as possible to improve reliability of results because the more similar results of repeats are, the more reliable the first outcome is.
12 of 14
Evaluating Methods and Results (reproducibility).
if compared results from different people are very similar, chances are there is a casual relationship therefore other scientists could gain results showing a similar relationship.
13 of 14
Evaluating Methods and Results (validity).
Ensure all variables were controlled appropriately to receive a result which answers the title of the investigation.
14 of 14
Other cards in this set
Card 2
Front
Anamalous results.
Back
Investigate, find out what happened and dismiss that result. No valid conclusions can be drawn from these type of results.
Comments
No comments have yet been made