## data collection -

data processing/analysis -

You are required to complete statistical measures. At this level of research, it

is simply not an option. The good news is that there is help for you. Below is a

Brief Course in Statistics written by Dru Martin at Ramstein High School. This

will assist you with your data analysis.

Data gathering + measurement:

Ultimately the analysis you can do, the graphs you can make and the conclusions

you can come to depend on the data that you gather. The richer in information

your measurements are, the better the rest of your project can be. To

illustrate:

is simply not an option. The good news is that there is help for you. Below is a

Brief Course in Statistics written by Dru Martin at Ramstein High School. This

will assist you with your data analysis.

Data gathering + measurement:

Ultimately the analysis you can do, the graphs you can make and the conclusions

you can come to depend on the data that you gather. The richer in information

your measurements are, the better the rest of your project can be. To

illustrate:

**1.**

The bacteria grew yes/noThe bacteria grew yes/no

**2.**

The bacteria grew (circle one) none some lotsThe bacteria grew (circle one) none some lots

**3.**

The bacteria covered ____% of the surface.The bacteria covered ____% of the surface.

**Number**

one is an example of a categorical measurement; your observation fits into one

of two categories. Number two is also a categorical measurement, but the

categories have an inherent order. Number three is a quantitative measurement.

As we move from 1 to 2 to 3, the amount of information contained in the data

increases, and the analysis possibilities increase as well. When possible, take

quantitative measurements of your dependent and independent

variables.

Graphing data: Your experiment will most likely be comparing a

control setup with an experimental setup. Therefore, your graphs should clearly

compare these two results, preferably on the same graph.

The type of graph you can draw depends on the type of data that you took. For

categorical data, a pie chart or histogram is best. For one quantitative measurement,

a bar graph, boxplot, stem-and-leaf diagram or dotplot are all good alternatives. If

you have a quantitative measurement for each of your experimental groups, you

will need to display all the groups on one graph. Finally, if you are looking at

the relationship between two quantitative measurements, a scatterplot is most

useful, and a least squares regression line may be

appropriate.

Graphing

data: Your experiment will most likely be comparing a control setup with an

experimental setup. Therefore, your graphs should clearly compare these two

results, preferably on the same graph.one is an example of a categorical measurement; your observation fits into one

of two categories. Number two is also a categorical measurement, but the

categories have an inherent order. Number three is a quantitative measurement.

As we move from 1 to 2 to 3, the amount of information contained in the data

increases, and the analysis possibilities increase as well. When possible, take

quantitative measurements of your dependent and independent

variables.

Graphing data: Your experiment will most likely be comparing a

control setup with an experimental setup. Therefore, your graphs should clearly

compare these two results, preferably on the same graph.

The type of graph you can draw depends on the type of data that you took. For

categorical data, a pie chart or histogram is best. For one quantitative measurement,

a bar graph, boxplot, stem-and-leaf diagram or dotplot are all good alternatives. If

you have a quantitative measurement for each of your experimental groups, you

will need to display all the groups on one graph. Finally, if you are looking at

the relationship between two quantitative measurements, a scatterplot is most

useful, and a least squares regression line may be

appropriate.

Graphing

data: Your experiment will most likely be comparing a control setup with an

experimental setup. Therefore, your graphs should clearly compare these two

results, preferably on the same graph.

**The**

type of graph you can draw depends on the type of data that you took. For

categorical data, a pie chart or histogram is best. For one quantitative

measurement, a bar graph, boxplot, stem-and-leaf diagram or dotplot are all good

alternatives. If you have a quantitative measurement for each of your

experimental groups, you will need to display all the groups on one graph.

Finally, if you are looking at the relationship between two quantitative

measurements, a scatterplot is most useful, and a least squares regression line

may be appropriate.

Video - Introduction to Statistical Analysis

OPTIONS FOR STATISTICAL ANALYSIS

T-Test PowerPoint Explanationtype of graph you can draw depends on the type of data that you took. For

categorical data, a pie chart or histogram is best. For one quantitative

measurement, a bar graph, boxplot, stem-and-leaf diagram or dotplot are all good

alternatives. If you have a quantitative measurement for each of your

experimental groups, you will need to display all the groups on one graph.

Finally, if you are looking at the relationship between two quantitative

measurements, a scatterplot is most useful, and a least squares regression line

may be appropriate.

Video - Introduction to Statistical Analysis

OPTIONS FOR STATISTICAL ANALYSIS

T-Test PowerPoint Explanation