Symphony Home Page | Tutorial | Use Symphony Now!

TUTORIAL

Symphony is a collection of several computer tools that allow you to gather and analyze data in order to investigate a scientific question. With this version of Symphony, you can gather data that has been collected by others and which is accessible via the Internet. This data gathering component of Symphony is called the DataWarehouse. You will be able to visualize the data that you gather by graphing it for analysis. We call the visualization component Viz-It. You will also be able to build models to try to understand how different systems work. The modeling component is called Model Builder (beta).

To start your investigation and to help you understand how to use Symphony, we will first guide you through the different tools in Symphony, using a sample question to start the investigating. After you see the different tools in Symphony, you will then begin investigating questions of your own.

Now that you are logged in, let’s start investigating a sample question. To get started, consider the following driving question: Are carbon monoxide levels worse in an urban area or an area with less population density?

How might you go about investigating this question? Here is a brief overview of the different activities you might do to investigate the question:

• Develop the problem.

• Gather the data you need for this problem.

• Visualize the data by building graphs of the data.

• Analyze the graphs in order to interpret what the graphs are "saying".

• Build models to try and better understand a system and data about that system.

• Review the problem to write down new questions that arise.

Given this overview of activities for your investigation, let’s see how you can use Symphony to perform these different activities.

Let’s start by looking at the Symphony main screen (figure 1). The main screen is divided into two areas:

Aside from the main screen, you can also bring up the Conductor window (figure 2). The Conductor window can give you hints about some of the big things that you can do next as you use Symphony. You can open the Conductor window by clicking on the conductor (we like to call him Watson) in the center of the main process map. Try it now.

Since you will be starting a new investigation, there is only one option in the Conductor window: Start my plan. Press this button and you will see a couple of things. First, you will notice some text in the Conductor window. The text will give you a brief description about the item in the button that you just pushed. Here, you will see some information about starting your plan.

Second, you will notice that the process map changed colors when you pressed the button in the Conductor window. Some of the Conductor options will color the map to give you some extra information. The colors on the map mean the following:

• A dark blue activity is the last activity that you performed and saved.

• Blue activities are recommended next steps.

You can bring up the Conductor at different points of this tour so you can see some of your different options.

Finally, there is one more item to note here: rollover help. If you move the mouse pointer over the different activities in the main process map, you will see a description of the selected activity pop up. Try this with the different process maps that you will see as we proceed with the Symphony tour.

We will start by developing the problem. Developing the problem involves describing your driving question. It also involves deciding on the factors involved in your question, describing your hypothesis, composing a general plan for your investigation, and recording research notes.

To set up an activity, you drag an activity button from a process "bubble" on the main process map into the Plan row of the Plan/Log grid. Let’s begin by setting up a Develop Problem activity in Step 1:

• Click and hold in the Develop Problem bubble until you see a Develop Problem button appear.

• Drag the Develop Problem button down and drop it into the Step 1 cell of the Plan row.

That’s it, you have now set up the first activity in your investigation. Notice that you now have a Develop Problem button in the Plan row. Pressing that button opens up a workspace in the activity area of the main screen where you can start developing the problem (figure 3).

Now that this workspace is open, do the following:

Notice now that the plan button for Develop Problem has moved down from the Plan row to the Log row. This means that you have worked on and saved the activity that was in the Plan row. Press the Put away button in the Develop Problem workspace to close it. If you now click on the dark button in the Log row, you’ll see that the Develop Problem workspace that you worked on and saved opens up. Throughout your investigation, you can open workspaces you’ve saved from your Log row to review or make additions to your previous work.

Now that you’ve worked on developing your problem, let’s move on to collecting data.

How will we go about collecting data? In this version of Symphony, we will search the DataWarehouse for data that pertains to the driving question. Recall the DataWarehouse contains data collected by others that you can access over the Internet.

To set up a Collect Data activity, you go through the same steps as you did to set up the previous activity. Click and hold in the Collect Data bubble of the process map until the Collect Data button appears and then drag it down to the Step 2 cell. Press the button you just added to the Plan row to open your data collection workspace (see figure 4).

Note that the data collection workspace is a bit different that the previous workspace you saw. The workspace opens up with a couple of items. First, you see a series of buttons that describe the different ways you could collect data. The text box next to these buttons gives a brief description of each data collection option. Currently, you can only search for data, but we want you to start seeing that there are different ways scientists can go about collecting data (marked by the other two choices).

To start working on data collection, click on the Search for data button. The workspace now shows some new items (figure 5):

The text box to the right of the buttons has changed to an area where you can reflect on and record your goals for this activity: why are you collecting data, what do you hope to gain, etc. The text box will contain prompts that describe some things you should think about here. Throughout Symphony, you will see these prompts to help you reflect on and focus on the work you should be doing in the different investigative activities.

So now, in the text box, explain your data collection goals. Recall the carbon monoxide driving question: What kind of data do you think you need to collect? Justify the type of data you hope to later obtain.

Next, notice the data search process map. This process map shows the steps needed to collect data from the DataWarehouse. You will search the DataWarehouse and the data that you find will be placed into a spreadsheet. Let’s start going through the data search process map:

Let’s take a look at the Michigan map. To select the city whose data you want, simply click on the name of the city (whose name will then appear in the box next to the map). Sometimes, in order to see the appropriate city, you might need to zoom in the map. You select the square area to zoom into by clicking and holding the left mouse button, dragging the mouse to make a box and releasing the mouse button. To zoom out, simply click on the map using the right mouse button. To move around the map (especially when you’ve zoomed in), click on the Navigator button. You will see a small map with a red box. Moving the red box changes your view on the map.

Select month from the first menu and average from the second menu (you might want to click on the menu just to see what the different options are. Click OK.

Select January 1, 1992 as the start date and January 1, 1994 as the end date. Click OK.

Notice that really long name in the Confirm window? That’s the name of the spreadsheet row where the data you find will be placed. The name is long because it explicitly defines the data you are looking for. You can change the name to something more reasonable. If you are satisfied with specifications you setup for the data search, click Yes. (If not, click No and go back the appropriate step in the map to make a change.)

Once you start the data search, Symphony will call the DataWarehouse to search for your data. The results appear in a spreadsheet, which you can go ahead and save, following the prompts on the screen.

Let’s add some more data to your spreadsheet. Make sure that the spreadsheet you just created is open. (If it isn’t, go to your spreadsheet list in the workspace, select the name of the spreadsheet and press Open.)

You have carbon monoxide data for a less dense area, now what kind of data should you look for? Let’s now add some data for a more urban area (say Livonia). Redo the steps from above, but this time add carbon monoxide data for Livonia.

Scroll through the spreadsheet to see the Livonia data. When you’ve added the new row of data to your spreadsheet, save the spreadsheet again.

Now that you’ve finished searching for data, there is one more area in the workspace to look at. Notice the What did you gain or learn in this activity? text box. Here, you can sum up what you did in this workspace so that you will have a record for this Step in your log. Write some notes describing what you did for later reference. (Again, as your investigations get bigger, you’ll want to have a good record of what exactly you did and why so you can show others the progress of your investigation).

Save the work in your data collection workspace (the Save my work button in the workspace) to your log and save your Symphony session (the Save button under the plan/log grid). Now that you have collected some data, let’s move on to the next step and start visualizing that data.

How will we go about visualizing data? We will use Viz-It to build graphs of the data that you have in your spreadsheets.

As you have been doing previously, drag a Visualize Data activity from the main process map to your Plan row and open the workspace (figure 6).

Click on the Build Graphs button and write some notes describing what graphs you want to build and why.

Let’s build some graphs. You will use Viz-It to graph the data in the spreadsheets you created earlier:

The window that opens has a menu that displays all of the different spreadsheets that you have built along with an area where the selected spreadsheet is displayed. Since you have only built one spreadsheet, the menu will only have one item. Select the spreadsheet name in the spreadsheet menu.

Now, you need to select the data you want to graph. Do this by clicking on the spreadsheet rows containing the data you want to visualize. Given the driving question, we want to graph Kalamazoo and Livonia carbon monoxide data, so click on both rows (for Kalamazoo and Livonia carbon monoxide). The rows that you select will turn red. Click This is what I want to close the window.

The resulting graph will now appear in a window (when you see the NetCharts splash screen, click on Press here to Continue to see the graph). The graph might be "zoomed-in" all the way. Click on the graph using your right mouse button to zoom out. (You can zoom in the graph the same way you zoomed in on the data collection Michigan map: by "drawing" a box using the left mouse button.) Also, if you move the mouse pointer on top of one of the data bars in the bar graph, a data value will popup showing the data value and date for the data point represented by the bar. (This effected is "turned on" by checking the checkbox in the Confirm Choices window. If no information pops up when you place the mouse pointer on a bar, you might not have checked this checkbox.)

Go ahead and save the graph, following the prompts on the screen. Type some notes describing what you accomplished and learned in the activity and save the workspace and the Symphony session.

Now that you have built a graph, let’s take a brief look at building models.

You will also be able to build system models using TheoryBuilder. You can set up a modeling activity just as you have set up data collection and graphing activities: drag a Model Data button from the process map to the plan grid and pressing the button to open a Build Model workspace.

TheoryBuilder is essentially a Java version of Model-It, so we will not go into detail about how to use TheoryBuilder. However, as a quick overview, we can briefly describe what you will be doing with TheoryBuilder. You can build models to describe how you think a system behaves. A system can be considered an environment where you will add objects. Each object can have attributes or factors. You can then set up relationships between the different attributes that describe how the different attributes affect other attributes. You can then see how the model you built behaves given the relationships that you have set up.

Let’s start wrapping up the Symphony tour.

The last thing to do is to revisit your driving question to see what you can say about the work you’ve completed so far.

Drag a Review Progress activity from the main process map to your Plan row. Open the Review Progress workspace (figure 7).

This workspace is where you can sum up the results you have generated so far to try to answer your driving question and to plan out what you need to do next. The different areas in the workspace show your driving question along with your original hypothesis and plan. Here, you should look at the results you have generated to try and support (or disprove) your original hypothesis.

You can look back at your results by pressing the List of things you built button. This will bring up a the Artifacts Log window (figure 8) that has buttons for every spreadsheet, graph, and model that you built. Pressing the buttons in the log window will bring up the specified item. This way, you can quickly bring up your different results as you try to answer the question in the workspace. Try it now by opening some of the things you have built on this tour.

Given your driving question and your hypothesis, try and write down a preliminary answer in text area. Does the work you’ve done so far support your hypothesis or does it dispute it? Do your have new questions as a result of what you’ve done. Remember, if you need to review or refer to any of your previous notes, you can always open the workspaces in your Log row.

As you try and write a preliminary answer, you might find that you need to revise your plan (for example, if you do not have enough evidence to support your hypothesis). Also, you might find that you need to change your hypothesis (for example, if your evidence disproves your hypothesis) and fix your plan to investigate the new hypothesis. Try and answer the different questions in the workspace and then save the workspace.

We are done for now. When you finish any Symphony session, put away any activity workspaces that you may have open, saving them as needed. Finally, save your plan/log grid by pressing the Save button under the grid in the main screen. To quit, simply press the Quit button under the plan/log grid and close your browser.

We have taken you through the different activities that you can do with Symphony. At this point, let’s consider a couple of points.

We ended our tour through Symphony at the Review Progress workspace. In this initial example, you might be able to provide some sort of answer for the driving question we gave you. However, in a real science investigation, rarely will you have a final answer so quickly. Most of the time, you will find that you need to do more work: you might need more data, more graphs, etc. Also, when your review your work, you might find that you are coming up with even more questions than answers.

This is the natural progression of a science investigation. You will find that an investigation follows an iterative process; that is, you go back and forth through the different activities over and over. And while our tour here took you through the different activities in a pretty linear order, you will find that you don’t always go through the different activities in a set order. Rather, you tend to "bounce around" the different activities.

So now it’s your turn. You’ve seen what you can do with Symphony, try exploring some questions of your own.