This test will mostly focus on statistics, I think, so you can skip most of the beginning~

Please download the printable version, as it has much prettier formatting!

Designing and Experiment

1. Focused problem/research question.

This is the section for telling the reader what your experiment was all about. A good research question should have:

-independent and dependent included

-species name included (if applicable)

-source of material (if applicable) ex: chocolate for M&M

When identifying relevant variables, choose the ones that would probably have an impact on your experiment if changed. For example, an experiment about the effects of temperature on melting ice doesn’t need to have the type of water as a variable. Only test one independent variable at a time!

Remember to include: independent, dependent, and controlled variables.

2. Controlling variables

Here, you should show how to prove that the variables are being controlled.

-indicate the range of independent variables. For example, 1, 3, 5, 7 drops of food coloring.

-indicate how you are measuring independent/dependent variables

-how are you controlling the controlled variables?

-all of this can be indicated in the variables or procedure section

3. Developing a method for collection of data

-have multiple trials/repetitions, this will let you have enough data to perform calculations

-aim for 15 points

-have a good range of independent variables

-measurement of dependent variable is relevant, for example, measuring oxygen production for rate of photosynthesis, not something random like number of leaves or heat of plant

Data Collection and Processing

1. Recording raw data

It’s as easy as it sounds! Simply make a table, and write down the quantitative and qualitative things you see with your experiment. There isn’t any error here except for systematic error.

2. Processing raw data

Make sure to include sample calculations for everything!

In biology, data processing revolves around statistics. Statistics is math that uses small samples of a population to draw conclusions about the larger population. The majority of statistics will be covered later.

Presenting Processed Data

Trend lines: use best fit lines to indicate direct relationships. When there’s continuous data, and both sets are quantitative

Bar graphs: use for disconnected data; or grouped data. You can use this when one of the data sets is qualitative.

Conclusion and Evaluation

1. Concluding

Here, state your conclusion with justifications based on your data and math. Make sure it has this stuff:

-Justification means research, along with citations of similar investigations to create supporting evidence

-Data should always be included in the conclusion

-Answer the question: Does your data support or refute your hypothesis. Remember, you can’t prove a hypothesis

-r values are used to show correlation between your independent and dependent variables. A quick review, an r value shows the strength of the correlation. You want this number to be close to 1.

2. Evaluating Procedures

Keep in mind you’re expected to evaluate weaknesses and limitations in this section.

Weakness: Possible mistakes made during your lab.

Limitations: Well, ‘limits’ to your lab. Things you couldn’t control, or if your results were limited to a certain temperature range, plant type, etc.

Make sure not to mention ‘human error.’ Be specific in what caused the error.

Evaluation helps determine if the weakness/limitation is relevant.

3. Improving the Investigation

Make it a goal to suggest an improvement for every weakness stated in the evaluating procedures section.

Statistics

There are three main parts of statistics:

Mean: the average. This gets skewed by outliers, so you have to assume normal distribution. The symbol for it is x ̅.

Mode: the most frequent number in a set. If all numbers appear only once, there is no mode.

Median: the number in the middle.

When your data is:

Symmetrical, you have a choice between all three.

Asymmetrical, use median

Two peaks, use mode.

Standard Deviation

Standard deviation measures the dispersion of data or the spread of data around the mean. So, the smaller the standard deviation, the closer the data is spread around the mean. The equation for standard deviation is:

√((∑▒〖(x-¯x)〗^2 )/((n-1)))

Where:

x = each data value

¯x = the mean or average

n = the number of values

∑= the sum across the values

You won’t have to solve the formula by hand, but you need to put this into every lab report to show as an example calculation. Make sure to include the key too, it describes what each variable does.

Typically:

68% of data lies in one standard deviation

95% of data lies in two standard deviations

99% of data lies in three standard deviations

This kind of data would be in a bell shape. These percentages will help you find outliers, so memorizing them will be helpful!

T-Tests

T-tests compare the difference between two means of different groups to see if that difference is important. Usually, T-tests are only useful if a sample size is greater than 10, and you’re only comparing two groups to determine a difference.

Use a calculator to find the calculated T-value. This is a lot like the formula for standard deviation, in that you only have to show the formula, not calculate it yourself.

T=(¯x-¯y)/√((〖S^2〗_x)/n_1 +(〖S^2〗_y)/n_2 )

Where:

X = mean of population1

Y = mean of population2

S = standard deviation

The larger the calculated t-value, the smaller the overlap in two data sets. Which means it’s more significant!

Finding calculated T-value with a calculator:

Input data into lists

STAT->TESTS->2-samp T Test

Enter information along with µ1≠ µ2

Polled = YES

Calculate!

Please download the printable version, as it has much prettier formatting!

Designing and Experiment

1. Focused problem/research question.

This is the section for telling the reader what your experiment was all about. A good research question should have:

-independent and dependent included

-species name included (if applicable)

-source of material (if applicable) ex: chocolate for M&M

When identifying relevant variables, choose the ones that would probably have an impact on your experiment if changed. For example, an experiment about the effects of temperature on melting ice doesn’t need to have the type of water as a variable. Only test one independent variable at a time!

Remember to include: independent, dependent, and controlled variables.

2. Controlling variables

Here, you should show how to prove that the variables are being controlled.

-indicate the range of independent variables. For example, 1, 3, 5, 7 drops of food coloring.

-indicate how you are measuring independent/dependent variables

-how are you controlling the controlled variables?

-all of this can be indicated in the variables or procedure section

3. Developing a method for collection of data

-have multiple trials/repetitions, this will let you have enough data to perform calculations

-aim for 15 points

-have a good range of independent variables

-measurement of dependent variable is relevant, for example, measuring oxygen production for rate of photosynthesis, not something random like number of leaves or heat of plant

Data Collection and Processing

1. Recording raw data

It’s as easy as it sounds! Simply make a table, and write down the quantitative and qualitative things you see with your experiment. There isn’t any error here except for systematic error.

2. Processing raw data

Make sure to include sample calculations for everything!

In biology, data processing revolves around statistics. Statistics is math that uses small samples of a population to draw conclusions about the larger population. The majority of statistics will be covered later.

Presenting Processed Data

Trend lines: use best fit lines to indicate direct relationships. When there’s continuous data, and both sets are quantitative

Bar graphs: use for disconnected data; or grouped data. You can use this when one of the data sets is qualitative.

Conclusion and Evaluation

1. Concluding

Here, state your conclusion with justifications based on your data and math. Make sure it has this stuff:

-Justification means research, along with citations of similar investigations to create supporting evidence

-Data should always be included in the conclusion

-Answer the question: Does your data support or refute your hypothesis. Remember, you can’t prove a hypothesis

-r values are used to show correlation between your independent and dependent variables. A quick review, an r value shows the strength of the correlation. You want this number to be close to 1.

2. Evaluating Procedures

Keep in mind you’re expected to evaluate weaknesses and limitations in this section.

Weakness: Possible mistakes made during your lab.

Limitations: Well, ‘limits’ to your lab. Things you couldn’t control, or if your results were limited to a certain temperature range, plant type, etc.

Make sure not to mention ‘human error.’ Be specific in what caused the error.

Evaluation helps determine if the weakness/limitation is relevant.

3. Improving the Investigation

Make it a goal to suggest an improvement for every weakness stated in the evaluating procedures section.

Statistics

There are three main parts of statistics:

Mean: the average. This gets skewed by outliers, so you have to assume normal distribution. The symbol for it is x ̅.

Mode: the most frequent number in a set. If all numbers appear only once, there is no mode.

Median: the number in the middle.

When your data is:

Symmetrical, you have a choice between all three.

Asymmetrical, use median

Two peaks, use mode.

Standard Deviation

Standard deviation measures the dispersion of data or the spread of data around the mean. So, the smaller the standard deviation, the closer the data is spread around the mean. The equation for standard deviation is:

√((∑▒〖(x-¯x)〗^2 )/((n-1)))

Where:

x = each data value

¯x = the mean or average

n = the number of values

∑= the sum across the values

You won’t have to solve the formula by hand, but you need to put this into every lab report to show as an example calculation. Make sure to include the key too, it describes what each variable does.

Typically:

68% of data lies in one standard deviation

95% of data lies in two standard deviations

99% of data lies in three standard deviations

This kind of data would be in a bell shape. These percentages will help you find outliers, so memorizing them will be helpful!

T-Tests

T-tests compare the difference between two means of different groups to see if that difference is important. Usually, T-tests are only useful if a sample size is greater than 10, and you’re only comparing two groups to determine a difference.

Use a calculator to find the calculated T-value. This is a lot like the formula for standard deviation, in that you only have to show the formula, not calculate it yourself.

T=(¯x-¯y)/√((〖S^2〗_x)/n_1 +(〖S^2〗_y)/n_2 )

Where:

X = mean of population1

Y = mean of population2

S = standard deviation

The larger the calculated t-value, the smaller the overlap in two data sets. Which means it’s more significant!

Finding calculated T-value with a calculator:

Input data into lists

STAT->TESTS->2-samp T Test

Enter information along with µ1≠ µ2

Polled = YES

Calculate!

- Attachments
- Scientific Method and Statistics Test.docx
*The much better formatted version!***You don't have permission to download attachments.**- (20 Kb) Downloaded 11 times