Creating a typical t-test report

A typical student t-test report can include tables with Group Statistics and Independent Samples Test results.

In the following sections we will build these tables using Qlik Senset-test functions applied to two independent groups of samples, Observation and Comparison. The corresponding tables for these samples would look like this:

Loading the sample data

Do the following:

1. Create a new app with a new sheet.

2. Enter the following in the data load editor:

   Table1:
   Crosstable (Type, Value)
   Load recno() as ID, * inline [
   Observation|Comparison
   35|2
   40|27
   12|38
   15|31
   21|1
   14|19
   46|1
   10|34
   28|3
   48|1
   16|2
   30|3
   32|2
   48|1
   31|2
   22|1
   12|3
   39|29
   19|37
   25|2 ] (delimiter is '|');

   In this load script, recno() is included because crosstable requires three arguments. So, recno() simply provides an extra argument, in this case an ID for each row. Without it, Comparison sample values would not be loaded.

3. Click to load data.

Creating the Group statistics table

Do the following:

1. In the data load editor, click to go to app view, and then click the sheet you created before.

   This opens the sheet view.

2. Click Edit sheet to edit the sheet.

3. From Charts, add a table, and from Fields, add Type as a dimension to the table.

4. Add the following expressions as measures.

   Example expressions

   Label	Expression
   N	Count(Value)
   Mean	Avg(Value)
   Standard Deviation	Stdev(Value)
   Standard Error Mean	Sterr(Value)

5. Click Sorting and make sure Type is at the top of the sorting list.

Result:

A Group statistics table for these samples would look like this:

Creating the Independent sample test table

Do the following:

1. Click Edit sheet to edit the sheet.

2. From Charts add a table with the following expression as a dimension to the table. =ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)) and give it the label Type.

3. Add the following expressions as measures:

   Example expressions

   Label	Expression
   conf	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_conf(Type, Value),TTest_conf(Type, Value, 0))
   t	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_t(Type, Value),TTest_t(Type, Value, 0))
   df	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_df(Type, Value),TTest_df(Type, Value, 0))
   Sig. (2-tailed)	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_sig(Type, Value),TTest_sig(Type, Value, 0))
   Mean Difference	TTest_dif(Type, Value)
   Standard Error Difference	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_sterr(Type, Value),TTest_sterr(Type, Value, 0))
   95% Confidence Interval (Lower)	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_lower(Type, Value,(1-(95)/100)/2),TTest_lower(Type, Value,(1-(95)/100)/2, 0))
   95% Confidence Interval (Upper)	if(ValueList (Dual('Equal Variance not Assumed', 0), Dual('Equal Variance Assumed', 1)),TTest_upper(Type, Value,(1-(95)/100)/2),TTest_upper(Type, Value,(1-(95)/100)/2, 0))

   Result:

   Independent sample test

   Type	conf	t	df	Sig. (2-tailed)	Mean Difference	Standard Error Difference	95% Confidence Interval (Lower)	95% Confidence Interval (Upper)
   Equal Variance not Assumed	0	3.534	37.116717335823	0.001	15.2	4.30101	6.48625	23.9137
   Equal Variance Assumed	8.706939	3.534	38	0.001	15.2	4.30101	6.49306	23.9069