Statistical Methods Descriptive Statistics Tabular graphical and numerical methods used to summarize or describe data Involves collecting presenting and characterizing data Inferential Statistics Process of using sample data to draw conclusions or make decisions about the characteristics of the population Two forms Estimation Hypothesis testing Examples of Inferential Statistics 1 A market researcher wishes to estimate the mean shift in attitude induced by a television advertisement based on a telephone survey of television viewers 2 A manufacturer has just received a large shipment of modified machine components She wishes to test whether the defective rate of the modified components does not exceed 1 of the defective rate of the original components Measures of Location Which is more sensitive to extreme values in the data Mean or Median Mean is more affected by extreme values outliers especially when sample size is small Median is less sensitive to extreme values Measures of Variability Association between two variables Thus far we have focuses on methods that are used to summarize the data for one variable at a time Often a manager is interested in the relationship between two variables Purchasing coffee and purchasing sugar 80 of the time customers purchase coffee and sugar together provide insights to supermarkets when they need to redesign their layouts Scatter Diagram Graphical representation of the relationship between two quantitative variables x and y Examples Numerical Measures of Association between Two Variables Covariance Measure of the linear association between two variables Positive values indicate a positive relationship Negative values indicate a negative relationship Sample Covariance Population Covariance Numerical Measures of Association between Two Variables Correlation Coefficient Unit free measure of degree of linear association between two variables Sample correlation coefficient Population correlation coefficient Charts Chart area all area of chart plot area titles legend and labels Plot area region containing the graphical representation of your data Everything you need is under Chart Tools Design Tab Format Tab Create a chart Steps Select the data source Select the chart type Add and format chart elements Size and position the chart chart visual representation of numerical data that compares data and helps reveal trends or patterns to help people make informed decisions x axis horizontal border y axis vertical border category axis descriptive group names or labels value axis displays incremental numbers to identify the worksheet values such as number of jobs or revenue used to create the chart legend key that identifies the color gradient picture texture or pattern assigned to each data series in a chart Selecting the Data Source Before creating a chart organize the worksheet data so that the values in columns are rows are on the same value system such as dollars or units make sure labels are descriptive and delete any blank rows or columns that exist in the primary data set Decide what you want to convey to your audience by answering these questions Does the worksheet hold a single set of data such as average snowfall at one ski resort or multiple sets of data such as average snowfall at several ski resorts Do you want to depict data for one specific time period or over several time periods such as several years or decades data point each cell containing a value data series a group of related data points that display in row s or column s in the worksheet category labels row and column labels such as job titles years growth etc Choosing a Chart Type different story Select a chart type that appropriately represents the data You can create different charts from the same data set but each chart tells a Column chart displays values in vertical columns where the height represents the value the taller the column the large the value Categories display along the vertical category axis Bar chart displays values in horizontal bars where the width represents the value the wider the bar the large the value Categories display along the vertical category axis Line chart displays category data on the horizontal axis and value data on the vertical axis Appropriate to show continuous data to depict trends over time such as months years or decades Pie chart shows proportion of individual data points to the sum of all those data points Steps to Create a Chart 1 Select the data and click the Quick Analysis button 2 Click CHARTS in the Quick Analysis gallery 3 Position the mouse over each recommended chart thumbnail to see the type of chart that would be created from the selected data 4 Click the thumbnail of the chart you want to create Another way to create a chart 1 Click the Insert tab 2 Click the chart type such as Column in the Charts group and click a chart subtype such as Clustered Column from the chart gallery 3 Click Recommended Charts in the Charts group to open the Insert Chart dialog box click a thumbnail of the chart you want then click OK Steps to Create a Column Chart Column chart displays data vertically in columns Create a column chart to compare values across different categories such as population among cities in a state or number of computer related jobs between two years Column charts are most effective when they are limited to seven or fewer categories If more categories exist the columns appear too close together making it difficult to read the labels Clustered column chart compares groups or clusters of columns set side by side for easy comparison Facilitates quick comparisons across data series and it is effective for comparing several data points among categories The row labels appear on the category axis and the yearly data series appear on columns with the value axis showing incremental numbers Stack column chart relationship of individual data points to the whole category Displays only one column for each category Used the stacked column chart when you want to compare total values across categories as well as to display the individual category values A disadvantage of the stacked column chart is that the segments without each column do not start at the same point making it more difficult to compare individual segment values across categories When you create a stacked column chart make sure data are additive each column represents a sum of the data for each segment Create a Bar Chart Bar chart compares values across categories using
View Full Document
Unlocking...