代做ECON10151: Computing for Social Scientists Lecture 1: Basic Data Analysis using Excel代做Python程序

日期：2024-12-12 05:13

Lecture 1: Basic Data Analysis using Excel

ECON10151: Computing for Social Scientists

September 22, 2024

In this lecture, we’ll introduce some key features of Excel that are especially useful when working with social science data. Excel has a variety of tools designed to make data analysis quicker and more efficient. Today, we’ll focus on essential functions and techniques that will help you clean and format datasets, create new variables, carry out basic statistical calculations, and generate simple visualisations.  By working through practical examples, you’ll gain hands-on experience, equipping you to apply these tools to your own data with confidence.

1    Data

The dataset we will be working with contains the average price and quantity of hand washing products sold in the UK each month during 2020. This data was sourced from the Kantar FMCG Purchase Panel.

For this exercise, we will assume:

•  The firm’s variable cost is £3 per unit of quantity.

Variable costs are expenses that fluctuate with production or sales volume. Common examples include:

–  Raw materials: The cost of materials required to produce each unit.

–  Direct labour: Wages paid to workers based on the number of units produced or hours worked.

–  Packaging: Costs associated with packaging each product.

•  The firm’s fixed cost is £8,500.

Fixed costs remain the same, regardless of production levels. Typical fixed costs include:

–  Rent payments: The cost of leasing a building or office space.

–  Insurance: Premiums paid for coverage, which do not vary with output.

–  Depreciation: The gradual reduction in the value of fixed assets like machinery or equipment.

Please download the spreadsheet titled ‘raw data’ from Blackboard to get started.

2    From Raw Data to Profit Analysis

We’ll follow a step-by-step approach to analyse the profit from sales.

2.1    Cleaning Data

Properly organising your data is the first crucial step in preparing it for analysis. Sometimes, we deal with imported or unstruc- tured data where multiple pieces of information are combined into a single cell, which needs to be split for proper analysis—just like the raw data we’re working with.

In Excel, the Text to Columns function is an effective data-splitting tool for separating data contained in one column into multiple columns, based on a specific delimiter (such as a comma, space, or tab). Here’s how to use it:

Step 1  Select the Data: Highlight the cells containing the text you want to split.

Step 2  Open ‘Text to Columns’: Go to the Data tab and select Text to Columns.

Step 3  Choose ‘Delimited’ : In the dialog box, choose Delimited, as our data is separated by commas.

Step 4  Set Delimiter: Check the Comma option as the delimiter (since the values in the data are comma-separated).

Step 5  Finish: Excel will show a preview of how the data will look. Once you’re happy with the result, click Finish.

After using Text to Columns to split the data, you might notice that some of the headers are spread across multiple columns.  In such cases, manually adjust the headers by combining the text and labelling them appropriately (e.g., “Quantity Sold (Litres)” or “Price per Litre”).  Ensure that each row of data aligns correctly with its respective column before proceeding with further  analysis.

2.2    Transposing Data

Sometimes your data is organised horizontally, but you need it vertically—or vice versa. For example, your headers may be in a row when you need them in a column, or data may be listed in columns but would be more useful in rows.

Transposing data in Excel means flipping the orientation, converting rows into columns or columns into rows. This can be very useful when the data’s current format doesn’t suit the analysis or visualisation you’re aiming for.

We’ll explore two methods to transpose data in Excel during this lecture.

1.  Paste Special Transpose:

Step 1  Select the Data: Highlight the range of cells you want to transpose, including headers if necessary.

Step 2  Copy the Data.

Step 3  Choose the Destination:  Click the cell, e.g. A7, where you want the transposed data to appear.  Ensure there’s enough space, as transposing will expand the data either vertically or horizontally.

Step 4  Paste Special: Navigate to Home tab → Paste → Paste Special, or right-click on the destination cell and choose Paste Special.

Step 5  Transpose the Data:  In the  Paste  Special dialog box, select Values (to paste only the values) and check the Transpose option at the bottom. Then click OK. Your data will now be flipped between rows and columns.

2.  TRANSPOSE Function:

The TRANSPOSE function is another way to transpose data, and it differs from the Paste Special method in that it creates a dynamic link. This means that if the original data changes, the transposed data will update automatically.

The syntax is:

= TRANSPOSE(array)

Here, the array represents the range of cells to be transposed.

Follow these steps:

Step 1  Choose the Destination: Again, make sure there’s enough space.

Step 2  Enter the Formula: In the selected range, type the formula

= TRANSPOSE(A1 : N4)

           where A1 : N4 is the range of your original data.

Step 3  Press Enter.

Note: If you’re using older versions of Excel, press ‘Ctrl + Shift + Enter’ (Windows) or ‘Command + Return ’ (Mac), as the TRANSPOSEfunction requires this to work as an array formula.

- Summary:

•  Use TRANSPOSE if you need a live link between your original and transposed data.  This is ideal for datasets that are regularly updated.

•  Use Paste Special when you need a one-time, static rearrangement of your data and don’t require it to update automati- cally.

2.3    Removing Duplicates

When importing data from external sources such as CSV files, databases, or surveys, it’s common to encounter duplicate entries. Redundant data can skew your analysis, so it’s essential to remove any duplicate values before proceeding.

Manually identifying and removing duplicate rows can be time-consuming, especially with large datasets.  Fortunately, Excel’s  Remove  Duplicates feature helps you quickly  find and eliminate duplicate entries, ensuring your dataset remains accurate and free of redundancy.

Follow these steps to remove duplicates in Excel:

Step 1  Select the Data Range: Highlight the range of cells where you want to remove duplicates. This can be a single column or an entire table, depending on what you need.

Step 2  Open the ‘Remove Duplicates’ Tool: Go to the Data tab and click on Remove Duplicates.

Step 3  Choose Columns to Check for Duplicates:  A dialog box will appear showing all the columns in the selected range. Tick or untick the boxes to specify which columns should be checked for duplicates.

•  If you select just one column, Excel will remove rows where the value in that column is repeated, even if other values in the row differ.

•  If you select multiple columns, Excel will treat rows as duplicates only if the combination of values in those columns is identical.

Step 4  Click OK: Once you’ve chosen the columns to check, click OK. Excel will display a message showing how many duplicates were found and removed, and how many unique entries remain.

- Important Notes:

•  Removing duplicates is permanent. Consider creating a backup of your dataset before using this tool to prevent accidental data loss.

•  The Remove Duplicates tool is case-insensitive, meaning it treats ‘Judith’ and ‘judith’ as duplicates.

2.4    Creating New Variables

Often, creating new variables is essential for deeper analysis. In this section, we will create three new variables: Total Revenue (TR), Total Cost (TC), and Profit. Additionally, we will categorise the data by creating a dummy variable and a string variable to indicate whether a profit was made.

1.  Calculating Total Revenue, Total Cost, and Profit

We will start by calculating Total Revenue (TR), Total Cost (TC), and Profit using simple Excel formulas. Step 1  Total Revenue (TR): This represents the total income from sales and is calculated as:

TR = Quantity Sold × Price

In Excel, you can enter this formula in a new column. For example, in cell E8, type:

= C8 * D8

where C8 contains the quantity sold, and D8 contains the price for January. Step 2  Total Cost (TC): Total Cost is the sum of variable and fixed costs:

TC = (Variable Cost ×Quantity Sold)+Fixed Cost

Recall that the variable cost is £3 per unit, and the fixed cost is £8,500. Enter these values in cells B21 and B22 respectively.

In Excel, you can enter the formula as follows in cell F8:

= (3*C8)+$B$22

Here, $B$22 locks both the column and row reference to cell B22, which contains the fixed cost (£8,500), ensuring that the reference cell doesn’t change when copying the formula down to other rows.

Step 3  Profit: Profit is calculated as the difference between Total Revenue and Total Cost:

Profit = TR−TC

In Excel, type the formula in cell G8 as:

= E8 − F8

where E8 contains TR, and F8 contains TC.

2.  Categorising Profit with a Dummy Variable

Next, we will categorise profit by creating a dummy variable.  A dummy variable is a numeric variable that takes the value 1 or 0. Here, we shall use 1 to represent a positive profit, and 0 will represent no profit or a loss. We’ll use Excel’s IF function to create this variable.

The IF function performs a conditional test and returns one value if the condition is TRUE and another if the condition is FALSE. The syntax is as follows:

= IF(logical_test,    value  if  true,    value  if  false)

where:

•  logical_test: The condition to check (e.g., G8 > 0 to test if profit is positive).

•  value_if_true: The value returned if the condition is TRUE (e.g., 1).

•  value_if_false: The value returned if the condition is FALSE (e.g., 0).

In this case, to create the dummy variable, type the following formula in a new column (e.g., cell H8): = IF(G8 > 0, 1, 0)

This formula assigns a value of 1 if the profit (in G8) is positive, and 0 if it is zero or negative.

3.  Creating a String Variable for Profit

To make the data more descriptive, we can create a string variable that labels whether a profit was made. Instead of using numeric values, we will use "YES" for profit and "NO" for no profit.

In a new column (e.g., I8), use the following formula:

= IF(G8 > 0, "YES","NO")

This formula will display "YES" if the profit is positive and "NO" if it is not.

Note: When using text values in Excel formulas, always enclose them in quotation marks ("   "). If you don’t, Excel will return an error (#NAME?).

2.4.1    Applying Formulas Across Rows Using the Fill Handle

Once you’ve entered your formulas, you can quickly apply them to the entire dataset using the Fill Handle. To do this:

•  Select the cell(s) containing the formula(s) you want to copy.

•  Hover over the bottom-right corner of the selected cell(s) until a small square appears—this is the Fill Handle.

•  Click and drag the Fill Handle down or across to fill the formula into the adjacent cells.

This method allows you to efficiently apply the formulas to all relevant rows or columns.

2.5    Conditional Formatting

Conditional formatting is a powerful tool in Excel that helps you visually emphasise specific values in your dataset.  In this exercise, we’ll use it to highlight the cells that display "YES" or "NO" for profit,making patterns easier to identify at a glance.

Step 1  Select the Range of Cells: Highlight the cells where you want to apply conditional formatting, e.g. I8:I19. Step 2  Open the Conditional Formatting Menu: Navigate to Home → Conditional Formatting.

Step 3  Create a Rule for "YES":

Select New Rule... from the drop-down menu, then choose the Classic style. Next, select Only format cells that contain from the list of rule types. In the options, choose Specific Text, select containing, and type YES in the text box.

Note: You can choose from a variety of pre-defined rules or create your own custom rule for more specific needs.

Step 4  Choose a Format:

In the Format with drop-down, select a fill colour to highlight the cells containing "YES" (e.g., Green  Fill with  Dark Green Text). Click OK to apply the rule.

Step 5  Create a Rule for "NO":

Repeat the process for the cells that contain "NO", but choose a different fill colour (e.g., Red Fill with Dark Red Text) to distinguish them from the "YES" entries.

2.6    Data Analysis Functions

To wrap up, we’ll use some of Excel’s built-in functions to summarise and analyse our data.  These functions will allow us to quickly count, calculate, and identify key metrics related to our profit data.

Step 1  Counting Profitable Months:

To count the number of months where profit was made, we can either use the COUNTIF function or sum the Profit dummy variable.

•  Using COUNTIF: This function counts the number of cells that meet a specific condition. To count the months with profit, use the following formula:

= COUNTIF(I8 : I19, "YES")

This counts the number of cells in the range I8:I19 that contain "YES".

•  Using SUM: If you’ve created a dummy variable for profit (1 for profit, 0 for no profit), you can use the SUM function to count the number of profitable months:

= SUM(H8 : H19)

This adds up the values in the dummy variable column (H8:H19), effectively counting how many months were profitable.

Step 2  Calculating the Average Profit:

To find the average profit, use the AVERAGE function. This will calculate the mean of the profit values in your dataset. Enter the following formula in a new cell:

= AVERAGE(G8 : G19)

This formula calculates the average profit from the values in the range G8:G19.

Step 3  Finding the Minimum and Maximum Profit:

You can quickly identify the lowest and highest profit using the MIN and MAX functions: = MIN(G8 : G19)

and

= MAX(G8 : G19)

These functions provide a quick and efficient way to summarise your data and gain insights into overall performance.

2.7    Plotting a Line Graph for Monthly Profit

Visualising data is a key part of analysis, as it allows you to easily identify trends and patterns. Lastly, we’ll create a line graph to display the monthly profit using Excel’s Insert tab.

Step 1  Select the Data: Highlight the range of cells containing the profit values, e.g., G7:G19, including the headers. Step 2  Open the Insert Tab: Go to the Insert tab, where you’ll find various charting and graphing options.

Step 3  Choose a Line Graph: Click the Line Chart icon and select the basic line chart option from the available choices. Step 4  Customising the Chart:

After the graph is created, you can format it to make it more informative and visually appealing:

•  Add Axis Titles: Navigate to Chart Design → Add Chart Element → Axis Titles, and then label the horizontal axis as “Month” and the vertical axis as “Profit” .

•  Title the Chart: Provide an appropriate title for the chart, e.g., Monthly Profit for 2020.

•  Customise the Line and Data Markers:  To customise the line’s  appearance, double-click on it to open the Format Data Series pane.  In the Fill & Line tab, click Marker, expand  Marker Options, and choose Built-in to select the desired marker type. You can also adjust the marker size for better visibility.

•  Fade Out Gridlines (Optional):  To  soften the gridlines, double-click on any gridline to select them.   In the Format Major Gridlines pane, under the Fill & Line tab, adjust the transparency level to your preference.

•  Add a Trendline: A trendline can represent the overall direction of the data over time. To add a trendline, go to Chart Design → Add Chart Element → Trendline, and select Linear.

In this example, a trendline with a positive slope is added, showing that monthly profits are trending upward.

To distinguish the trendline from the data line, you can add a legend: Go to Chart Design → Add Chart Element → Legend, and select the desired location (e.g., Bottom).

This process offers a simple way to visualise monthly profit using a line graph, making it easier to spot trends and patterns over time. Now that you’ve learned how to create and customise a line graph, try experimenting with other types of charts, such as bar or pie charts, to visualise different aspects of your data on your own.