lib/statsample/test/t.rb in statsample-0.9.0 vs lib/statsample/test/t.rb in statsample-0.10.0

@@ -29,11 +29,11 @@
           else
             den=sqrt((s1**2).quo(n1) + (s2**2).quo(n2))
           end
           num.quo(den)
         end
-         # Degrees of freedom for equal variance
+        # Degrees of freedom for equal variance on t test
         def df_equal_variance(n1,n2)
           n1+n2-2
         end
         # Degrees of freedom for unequal variance
         # * <tt>s1</tt>: sample 1 standard deviation
@@ -65,28 +65,21 @@
       #  t = -1.1573, p=0.2474, d.f=999
 
       class OneSample
         include Math
         include Statsample::Test
-        include DirtyMemoize
         # Options
         attr_accessor :opts
         # Name of test
         attr_accessor :name
         # Population mean to contrast
         attr_accessor :u
         # Degress of freedom
         attr_reader :df
-        # Value of t
-        attr_reader :t
-        # Probability
-        attr_reader :probability
         # Tails for probability (:both, :left or :right)
         attr_accessor :tails 
         
-        dirty_writer :u, :tails
-        dirty_memoize :t, :probability
         # Create a One Sample T Test
         # Options:
         # * :u = Mean to compare. Default= 0
         # * :name = Name of the analysis
         # * :tails = Tail for probability. Could be :both, :left, :right
@@ -98,18 +91,18 @@
           @u=@opts[:u]
           @tails=@opts[:tails]
           @df= @vector.n_valid-1
           @t=nil
         end        
-       
+        def t
+          T.one_sample(@vector.mean, @u, @vector.sd, @vector.n_valid)
+        end
         
-        # Set t and probability for given u
-        def compute
-          @t  = T.one_sample(@vector.mean, @u, @vector.sd, @vector.n_valid)
-          @probability = p_using_cdf(Distribution::T.cdf(@t, @df), tails)
+        def probability
+          p_using_cdf(Distribution::T.cdf(t, @df), tails)
         end
-        # Presents summary of analysis
+        # Summary of analysis
         # 
         def summary
           ReportBuilder.new(:no_title=>true).add(self).to_text
         end
         def report_building(b) # :nodoc:
@@ -152,10 +145,12 @@
 
       class TwoSamplesIndependent
         include Math
         include Statsample::Test
         include DirtyMemoize
+        include GetText
+        bindtextdomain("statsample")
         # Options
         attr_accessor :opts
         # Name of test
         attr_accessor :name
         # Degress of freedom (equal variance)
@@ -202,28 +197,39 @@
           @probability_equal_variance = p_using_cdf(Distribution::T.cdf(@t_equal_variance, @df_equal_variance), tails)
           
           @probability_not_equal_variance = p_using_cdf(Distribution::T.cdf(@t_not_equal_variance, @df_not_equal_variance), tails)
 
         end
+        # Cohen's d is a measure of effect size. Its defined as the difference between two means divided by a standard deviation for the data
+        def d
+          n1=@v1.n_valid
+          n2=@v2.n_valid
+          num=@[e-mail]
+          den=Math::sqrt( ((n1-1)*@v1.sd+(n2-1)*@v2.sd).quo(n1+n2))
+          num.quo(den)
+        end
+        
         # Presents summary of analysis
-        # 
         def summary
           ReportBuilder.new(:no_title=>true).add(self).to_text
         end
+        
         def report_building(b) # :nodoc:
           b.section(:name=>@name) {|g|
-            g.table(:name=>"Mean and standard deviation", :header=>["Variable", "m", "sd","n"]) {|t|
-              t.row([1,"%0.4f" % @v1.mean,"%0.4f" % @v1.sd,@v1.n_valid])
-              t.row([2,"%0.4f" % @v2.mean,"%0.4f" % @v2.sd, @v2.n_valid])
+            g.table(:name=>_("Mean and standard deviation"), :header=>["Variable", "m", "sd","n"]) {|t|
+              t.row([@v1.name,"%0.4f" % @v1.mean,"%0.4f" % @v1.sd,@v1.n_valid])
+              t.row([@v2.name,"%0.4f" % @v2.mean,"%0.4f" % @v2.sd, @v2.n_valid])
             }
-            g.section(:name=>"Levene Test") {|g1|
-              g1.parse_element(Statsample::Test.levene([@v1,@v2]))
-            }
+            g.parse_element(Statsample::Test.levene([@v1,@v2],:name=>_("Levene test for equality of variances")))
             
-            g.table(:name=>"T statistics",:header=>["Type","t","df", "p (#{tails} tails)"]) {|t|
-              t.row(["Equal variance", "%0.4f" % t_equal_variance, df_equal_variance, "%0.4f" % probability_equal_variance])
-              t.row(["Non equal variance", "%0.4f" % t_not_equal_variance, "%0.4f" % df_not_equal_variance, "%0.4f" % probability_not_equal_variance])
+            g.table(:name=>_("T statistics"),:header=>["Type","t","df", "p (#{tails} tails)"].map{|v| _(v)}) {|t|
+              t.row([_("Equal variance"), "%0.4f" % t_equal_variance, df_equal_variance, "%0.4f" % probability_equal_variance])
+              t.row([_("Non equal variance"), "%0.4f" % t_not_equal_variance, "%0.4f" % df_not_equal_variance, "%0.4f" % probability_not_equal_variance])
             }
+            g.table(:name=>_("Effect size")) do |t|
+              t.row ['x1-x2', "%0.4f" % (@[e-mail])]
+              t.row ['d', "%0.4f" % d]
+            end
           }
         end
       end      
     end
   end