Data

​Data: Patterns, errors, issues, ethics

By Elizabeth Kuzara

​There are some things we need to know about data, and how it is collected to be able to appropriately use and analyze it. As nice looking as any graph is, or as convincing as the description, we must still critically evaluate our data.

​Accuracy vs Precision

​Accuracy is our ability to provide and measure correct data. If we have a scale that tells us that a paper clip weights 3lbs, we know that scale is not accurate.

If that same scale says that all of our paperclips are approximately 3lbs we could say it is precise, as in it is giving data within the same range for the reading.

So precision can be described as the reliability of a reading or measure.

​A

​B

Which scale would be more precise? Scale A or Scale B?

​Precision also relates to how refined our data is. If a scale reads down to thousandths of a gram it is more precise than one that only shows tenths of a gram.

In your notes write these to values and label high and low precision.

​High precision
0.02356g

Low precision
0.0g

​In these two readings, we have a scale that can provide a reading to the hundrend-thousandths place

This second reading only goes to the the tenths place. The sample is so small that the reading from this scale cannot detect the sample.

Multiple Choice

A scale that weighs an apple shows that it weight 20lbs, but each time it is weighed it has the same reading. What can we say about this scale.

• It is accurate and precise

• It is inaccurate and not precise

• It is precise and inaccurate

• It is not precise and it is not accurate

​Outliers

​When we are collecting and entering data it is possible to make errors. When we evaluate data and there is a data point(s) that clearly do not belong to the set, we call that an OUTLIER. When we review and clean up data, we may need to make a choice to keep or remove that data.

​On the graph in your notes circle the outlier.

​Trends and Patterns

​When we evaluate data we are often looking for trends or patterns. Trends are the direction of the data. Usually an upward trend is an increase and a downward trend is a decrease.

​On your graph draw and label and upward (increasing) trend and in another color, a downward (decreasing) trend.

​Patterns

​Patterns in data may allow us to come to conclusions about cycles. Such as migration, mating, predation, and seasons.

​Correlation and Causation

​Correlation is how the data is related. Causation is if one caused the other. For example, the predator and prey population changes due to the changes in each others populations.

​Although the data here follows a similar pattern, it is probably not likely that Birds of Prey influenced anyone to become a firefighter.

​Conclusions

​Data should allow us to draw a conclusion, and our conclusions should follow the CER model. Claim - what you're stating about the data
Evidence - what the evidence shows
Reasoning - what the data and the research does to support your claim

​On the next slide, evaluate the statement and the graph and write a CER summary.

​Due to declining populations all wild California Condors were brought into captivity. The keepers kept the birds safe and ensured that their offspring survived. Over time they have released and monitored California Condors very carefully.

​Credibility and Ethics

​We must ensure that the data and evidence we read and believe comes from credible sources and is ethically produced.
Credibility - able to be believed due to expertise and reliable and reviewed research and work.
Ethics - Honesty, integrity, and reduction or mitigation of harm in the scientific process. In the scientific process we want to avoid causing undue harm or misrepresenting our findings for personal, political, or capitalistic gain.