statsample 0.6.5 → 0.6.7

data/lib/statsample/reliability.rb

@@ -1,28 +1,28 @@
 module Statsample
   module Reliability
-	  class << self
+    class << self
       # Calculate Chonbach's alpha for a given dataset.
       # only uses tuples without missing data
-			def cronbach_alpha(ods)
-				ds=ods.dup_only_valid
-				n_items=ds.fields.size
-				sum_var_items=ds.vectors.inject(0) {|ac,v|
-          ac+v[1].variance_sample }
-				total=ds.vector_sum
-				(n_items / (n_items-1).to_f) * (1-(sum_var_items/ total.variance_sample))
-			end
+      def cronbach_alpha(ods)
+        ds=ods.dup_only_valid
+        n_items=ds.fields.size
+        sum_var_items=ds.vectors.inject(0) {|ac,v|
+        ac+v[1].variance_sample }
+        total=ds.vector_sum
+        (n_items / (n_items-1).to_f) * (1-(sum_var_items/ total.variance_sample))
+      end
       # Calculate Chonbach's alpha for a given dataset
       # using standarized values for every vector.
       # Only uses tuples without missing data
-      
+
       def cronbach_alpha_standarized(ods)
         ds=ods.dup_only_valid.fields.inject({}){|a,f|
           a[f]=ods[f].vector_standarized; a
         }.to_dataset
         cronbach_alpha(ds)
       end
-		end   
-		class ItemCharacteristicCurve
+    end
+    class ItemCharacteristicCurve
       attr_reader :totals, :counts,:vector_total
       def initialize (ds, vector_total=nil)
         vector_total||=ds.vector_sum
@@ -39,28 +39,28 @@ module Statsample
           tot=@vector_total[i]
           @totals[tot]||=0
           @totals[tot]+=1
-					@ds.fields.each  do |f|
+          @ds.fields.each  do |f|
             item=row[f].to_s
             @counts[f][tot]||={}
             @counts[f][tot][item]||=0
-            @counts[f][tot][item] += 1 
+            @counts[f][tot][item] += 1
           end
-					i+=1
+          i+=1
         end
       end
       def curve_field(field, item)
         out={}
         item=item.to_s
         @totals.each{|value,n|
-            count_value= @counts[field][value][item].nil? ? 0 : @counts[field][value][item] 
-            out[value]=count_value.to_f/n.to_f
+          count_value= @counts[field][value][item].nil? ? 0 : @counts[field][value][item]
+          out[value]=count_value.to_f/n.to_f
         }
         out
       end
     end
-		class ItemAnalysis
+    class ItemAnalysis
       attr_reader :mean, :sd,:valid_n, :alpha , :alpha_standarized
-			def initialize(ds)
+      def initialize(ds)
         @ds=ds.dup_only_valid
         @total=@ds.vector_sum
         @item_mean=@ds.vector_mean.mean
@@ -74,11 +74,11 @@ module Statsample
           @alpha = Statsample::Reliability.cronbach_alpha(ds)
           @alpha_standarized = Statsample::Reliability.cronbach_alpha_standarized(ds)
         rescue => e
-          raise DatasetException.new(@ds,e), "Problem on calculate alpha" 
+          raise DatasetException.new(@ds,e), "Problem on calculate alpha"
         end
-			end
-            # Returns a hash with structure
-			def item_characteristic_curve
+      end
+      # Returns a hash with structure
+      def item_characteristic_curve
         i=0
         out={}
         total={}
@@ -89,7 +89,7 @@ module Statsample
             total[f]||={}
             out[f][tot]||= 0
             total[f][tot]||=0
-            out[f][tot]+= row[f] 
+            out[f][tot]+= row[f]
             total[f][tot]+=1
           end
           i+=1
@@ -100,79 +100,77 @@ module Statsample
           end
         end
         out
-			end
+      end
       def gnuplot_item_characteristic_curve(directory, base="crd",options={})
         require 'gnuplot'
-        
+
         crd=item_characteristic_curve
-        @ds.fields.each {|f|
-            x=[]
-            y=[]
-        Gnuplot.open do |gp|
-        Gnuplot::Plot.new( gp ) do |plot|
-            crd[f].sort.each{|tot,prop|
-               x.push(tot)
-               y.push((prop*100).to_i.to_f/100)
-           }
-        plot.data << Gnuplot::DataSet.new( [x, y] ) do |ds|
-        ds.with = "linespoints"
-        ds.notitle
-        end
-        
-        end
+        @ds.fields.each  do |f|
+          x=[]
+          y=[]
+          Gnuplot.open do |gp|
+            Gnuplot::Plot.new( gp ) do |plot|
+              crd[f].sort.each do |tot,prop|
+                x.push(tot)
+                y.push((prop*100).to_i.to_f/100)
+              end
+              plot.data << Gnuplot::DataSet.new( [x, y] ) do |ds|
+                ds.with = "linespoints"
+                ds.notitle
+              end
+
+            end
+          end
         end
-        }
-        
       end
       def svggraph_item_characteristic_curve(directory, base="icc",options={})
         require 'statsample/graph/svggraph'
         crd=ItemCharacteristicCurve.new(@ds)
-        @ds.fields.each {|f|
-         factors=@ds[f].factors.sort
-         options={
-                 :height=>500,
-                 :width=>800,
-                 :key=>true
-         }.update(options)
-         graph = ::SVG::Graph::Plot.new(options)
-         factors.each{|factor|
-             factor=factor.to_s
-             dataset=[]
-                 crd.curve_field(f, factor).each{|tot,prop|
-                     dataset.push(tot)
-                     dataset.push((prop*100).to_i.to_f/100)
-                  }
-              graph.add_data({
-                      :title=>"#{factor}",
-                     :data=>dataset
-              })
-         }
-         File.open(directory+"/"+base+"_#{f}.svg","w") {|fp|
-             fp.puts(graph.burn())
-         }
-        }
-         
+        @ds.fields.each do |f|
+          factors=@ds[f].factors.sort
+          options={
+            :height=>500,
+            :width=>800,
+            :key=>true
+          }.update(options)
+          graph = ::SVG::Graph::Plot.new(options)
+          factors.each do |factor|
+            factor=factor.to_s
+            dataset=[]
+            crd.curve_field(f, factor).each do |tot,prop|
+              dataset.push(tot)
+              dataset.push((prop*100).to_i.to_f/100)
+            end
+            graph.add_data({
+              :title=>"#{factor}",
+              :data=>dataset
+            })
+          end
+          File.open(directory+"/"+base+"_#{f}.svg","w") {|fp|
+            fp.puts(graph.burn())
+          }
+        end
       end
-			def item_total_correlation
-				@ds.fields.inject({}) do |a,v|
-					vector=@ds[v].dup
-					ds2=@ds.dup
-					ds2.delete_vector(v)
-					total=ds2.vector_sum
-					a[v]=Statsample::Bivariate.pearson(vector,total)
-					a
+      def item_total_correlation
+        @ds.fields.inject({}) do |a,v|
+          vector=@ds[v].dup
+          ds2=@ds.dup
+          ds2.delete_vector(v)
+          total=ds2.vector_sum
+          a[v]=Statsample::Bivariate.pearson(vector,total)
+          a
         end
-			end
-			def item_statistics
-				@ds.fields.inject({}) do |a,v|
-					a[v]={:mean=>@ds[v].mean,:sds=>@ds[v].sds}
-					a
-				end
-			end
+      end
+      def item_statistics
+        @ds.fields.inject({}) do |a,v|
+          a[v]={:mean=>@ds[v].mean,:sds=>@ds[v].sds}
+          a
+        end
+      end
       # Returns a dataset with cases ordered by score
       # and variables ordered by difficulty
-      
-			def item_difficulty_analysis
+
+      def item_difficulty_analysis
         dif={}
         @ds.fields.each{|f| dif[f]=@ds[f].mean }
         dif_sort=dif.sort{|a,b| -(a[1]<=>b[1])}
@@ -190,20 +188,20 @@ module Statsample
         ds_new.update_valid_data
         ds_new
       end
-			def stats_if_deleted
-				@ds.fields.inject({}) do |a,v|
-					ds2=@ds.dup
-					ds2.delete_vector(v)
-					total=ds2.vector_sum
-					a[v]={}
-					a[v][:mean]=total.mean
-					a[v][:sds]=total.sds
-					a[v][:variance_sample]=total.variance_sample
-					a[v][:alpha]=Statsample::Reliability.cronbach_alpha(ds2)
-					a
+      def stats_if_deleted
+        @ds.fields.inject({}) do |a,v|
+          ds2=@ds.dup
+          ds2.delete_vector(v)
+          total=ds2.vector_sum
+          a[v]={}
+          a[v][:mean]=total.mean
+          a[v][:sds]=total.sds
+          a[v][:variance_sample]=total.variance_sample
+          a[v][:alpha]=Statsample::Reliability.cronbach_alpha(ds2)
+          a
         end
-			end
-			def html_summary
+      end
+      def html_summary
 				html = <<EOF
 <p><strong>Summary for scale:</strong></p>
 <ul>
@@ -229,27 +227,26 @@ Correl.</th><th>Alpha if
 deleted</th></thead>
 EOF
 
-itc=item_total_correlation
-sid=stats_if_deleted
-is=item_statistics
-@ds.fields.each {|f|
-	html << <<EOF
-<tr>
-	<td>#{f}</td>
-    <td>#{sprintf("%0.5f",is[f][:mean])}</td>
-    <td>#{sprintf("%0.5f",is[f][:sds])}</td>
-	<td>#{sprintf("%0.5f",sid[f][:mean])}</td>
-	<td>#{sprintf("%0.5f",sid[f][:variance_sample])}</td>
-	<td>#{sprintf("%0.5f",sid[f][:sds])}</td>
-	<td>#{sprintf("%0.5f",itc[f])}</td>
-	<td>#{sprintf("%0.5f",sid[f][:alpha])}</td>
-</tr>
+        itc=item_total_correlation
+        sid=stats_if_deleted
+        is=item_statistics
+        @ds.fields.each {|f|
+          html << <<EOF
+          <tr>
+          <td>#{f}</td>
+          <td>#{sprintf("%0.5f",is[f][:mean])}</td>
+          <td>#{sprintf("%0.5f",is[f][:sds])}</td>
+          <td>#{sprintf("%0.5f",sid[f][:mean])}</td>
+          <td>#{sprintf("%0.5f",sid[f][:variance_sample])}</td>
+          <td>#{sprintf("%0.5f",sid[f][:sds])}</td>
+          <td>#{sprintf("%0.5f",itc[f])}</td>
+          <td>#{sprintf("%0.5f",sid[f][:alpha])}</td>
+          </tr>
 EOF
-}
-html << "</table><hr />"
-html
-			end
-		end		
-		
-	end
+        }
+        html << "</table><hr />"
+        html
+      end
+    end
+  end
 end

data/lib/statsample/srs.rb

@@ -2,161 +2,164 @@ module Statsample
 	# Several methods to estimate parameters for simple random sampling
 	module SRS
 		class << self
-			
-        ########################
-        #
-        # Proportion estimation
-        #
-        ########################
-        # Finite population correction (over variance)
-        # Source: Cochran(1972)
-        def fpc_var(sam,pop)
-            (pop - sam).quo(pop - 1)
-        end
-        # Finite population correction (over standard deviation)
-        def fpc(sam,pop)
-            Math::sqrt((pop-sam).quo(pop-1))
-        end
-        
-        # Non sample fraction.
-        #
-        # 1 - sample fraction
-        def qf(sam , pop)
-            1-(sam.quo(pop))
-        end
-        # Sample size estimation for proportions, infinite poblation
-        def estimation_n0(d,prop,margin=0.95)
-            t=Distribution::Normal.p_value(1-(1-margin).quo(2))
-            var=prop*(1-prop)
-            t**2*var.quo(d**2)
-        end
-        # Sample size estimation for proportions, finite poblation.
-        def estimation_n(d,prop,n_pobl,margin=0.95)
-            n0=estimation_n0(d,prop,margin)
-            n0.quo( 1 + ((n0 - 1).quo(n_pobl)))
-        end
-        
-        
-        # Proportion confidence interval with t values
-        # Uses estimated proportion, sample without replacement.
-        
-        def proportion_confidence_interval_t(prop, n_sample, n_population, margin=0.95)
-            t = Distribution::T.p_value(1-((1-margin).quo(2)) , n_sample-1)
-            proportion_confidence_interval(prop,n_sample,n_population, t)
-        end
-        
-        # Proportion confidence interval with z values
-        # Uses estimated proportion, sample without replacement.
-        def proportion_confidence_interval_z(p, n_sample, n_population, margin=0.95)
-            z=Distribution::Normal.p_value(1-((1-margin).quo(2)))
-            proportion_confidence_interval(p,n_sample,n_population, z)
-        end
-        # Proportion confidence interval with x value
-        # Uses estimated proportion, sample without replacement
-        
-        def proportion_confidence_interval(p, sam,pop , x)
-            f=sam.quo(pop)
-            one_range=x * Math::sqrt((qf(sam, pop) * p * (1-p)).quo(sam-1)) + (1.quo(sam * 2.0))
-            [p-one_range, p+one_range]
-        end
-        # Standard deviation for sample distribution of a proportion
-        # Know proportion, sample with replacement.
-        # Based on http://stattrek.com/Lesson6/SRS.aspx
-        def proportion_sd_kp_wr(p, n_sample)
-            Math::sqrt(p*(1-p).quo(n_sample))
-        end
-        # Standard deviation for sample distribution of a proportion
-        # Know proportion, sample without replacement.
-        #
-        # Sources: 
-        # * http://stattrek.com/Lesson6/SRS.aspx
-        # * Cochran(1972)
-        def proportion_sd_kp_wor(p, sam, pop)
-            fpc(sam,pop)*Math::sqrt(p*(1-p).quo(sam))
-        end
-        # Standard deviation for sample distribution of a proportion
-        # Estimated proportion, sample with replacement
-        # Based on http://stattrek.com/Lesson6/SRS.aspx.
-        def proportion_sd_ep_wr(p, n_sample)
-            Math::sqrt(p*(1-p).quo(n_sample-1))
-        end                                       
-        # Standard deviation for sample distribution of a proportion.
-        # Estimated proportion, sample without replacement.
-        # Source: Cochran, 1972, Técnicas de muestreo
-        def proportion_sd_ep_wor(p, sam,pop)
-            fsc=(pop-sam).quo((sam-1)*pop)
-            Math::sqrt(fsc*p*(1-p))
-        end
-        
-         # Total estimation sd based on sample. 
-        # Known proportion, sample without replacement
-        # Source: Cochran(1972)
-        def proportion_total_sd_kp_wor(prop, sam, pop)
-            pob * proportion_sd_kp_wor(p, sam, pop)
-        end
-        # Total estimation sd based on sample. 
-        # Estimated proportion, sample without replacement
-        # Source: Cochran(1972)
-        def proportion_total_sd_ep_wor(prop, sam, pop)
-            fsc=((pop - sam).to_f / ( sam - 1))
-            Math::sqrt(fsc*pop*prop*(1-prop))
-        end 
-        
-        ########################
-        #
-        # Mean stimation
-        #
-        ########################
+      ########################
+      #
+      # :SECTION: Proportion estimation
+      #
+      # Function for estimation of proportions
+      ########################
+      
+      #
+      # Finite population correction (over variance)
+      # Source: Cochran(1972)
+      def fpc_var(sam,pop)
+          (pop - sam).quo(pop - 1)
+      end
+      # Finite population correction (over standard deviation)
+      def fpc(sam,pop)
+          Math::sqrt((pop-sam).quo(pop-1))
+      end
+      
+      # Non sample fraction.
+      #
+      # 1 - sample fraction
+      def qf(sam , pop)
+          1-(sam.quo(pop))
+      end
+      # Sample size estimation for proportions, infinite poblation
+      def estimation_n0(d,prop,margin=0.95)
+          t=Distribution::Normal.p_value(1-(1-margin).quo(2))
+          var=prop*(1-prop)
+          t**2*var.quo(d**2)
+      end
+      # Sample size estimation for proportions, finite poblation.
+      def estimation_n(d,prop,n_pobl,margin=0.95)
+          n0=estimation_n0(d,prop,margin)
+          n0.quo( 1 + ((n0 - 1).quo(n_pobl)))
+      end
+      
+      
+      # Proportion confidence interval with t values
+      # Uses estimated proportion, sample without replacement.
+      
+      def proportion_confidence_interval_t(prop, n_sample, n_population, margin=0.95)
+          t = Distribution::T.p_value(1-((1-margin).quo(2)) , n_sample-1)
+          proportion_confidence_interval(prop,n_sample,n_population, t)
+      end
+      
+      # Proportion confidence interval with z values
+      # Uses estimated proportion, sample without replacement.
+      def proportion_confidence_interval_z(p, n_sample, n_population, margin=0.95)
+          z=Distribution::Normal.p_value(1-((1-margin).quo(2)))
+          proportion_confidence_interval(p,n_sample,n_population, z)
+      end
+      # Proportion confidence interval with x value
+      # Uses estimated proportion, sample without replacement
+      
+      def proportion_confidence_interval(p, sam,pop , x)
+          f=sam.quo(pop)
+          one_range=x * Math::sqrt((qf(sam, pop) * p * (1-p)).quo(sam-1)) + (1.quo(sam * 2.0))
+          [p-one_range, p+one_range]
+      end
+      # Standard deviation for sample distribution of a proportion
+      # Know proportion, sample with replacement.
+      # Based on http://stattrek.com/Lesson6/SRS.aspx
+      def proportion_sd_kp_wr(p, n_sample)
+          Math::sqrt(p*(1-p).quo(n_sample))
+      end
+      # Standard deviation for sample distribution of a proportion
+      # Know proportion, sample without replacement.
+      #
+      # Sources: 
+      # * http://stattrek.com/Lesson6/SRS.aspx
+      # * Cochran(1972)
+      def proportion_sd_kp_wor(p, sam, pop)
+          fpc(sam,pop)*Math::sqrt(p*(1-p).quo(sam))
+      end
+      # Standard deviation for sample distribution of a proportion
+      # Estimated proportion, sample with replacement
+      # Based on http://stattrek.com/Lesson6/SRS.aspx.
+      def proportion_sd_ep_wr(p, n_sample)
+          Math::sqrt(p*(1-p).quo(n_sample-1))
+      end                                       
+      # Standard deviation for sample distribution of a proportion.
+      # Estimated proportion, sample without replacement.
+      # Reference: 
+      # * Cochran, 1972, Técnicas de muestreo
+      def proportion_sd_ep_wor(p, sam,pop)
+          fsc=(pop-sam).quo((sam-1)*pop)
+          Math::sqrt(fsc*p*(1-p))
+      end
+      
+      # Total estimation sd based on sample. 
+      # Known proportion, sample without replacement
+      # Reference: 
+      # * Cochran(1972)
+      def proportion_total_sd_kp_wor(prop, sam, pop)
+          pob * proportion_sd_kp_wor(p, sam, pop)
+      end
+      # Total estimation sd based on sample. 
+      # Estimated proportion, sample without replacement
+      # Source: Cochran(1972)
+      def proportion_total_sd_ep_wor(prop, sam, pop)
+          fsc=((pop - sam).to_f / ( sam - 1))
+          Math::sqrt(fsc*pop*prop*(1-prop))
+      end 
+      
+      ########################
+      #
+      # :SECTION:  Mean stimation
+      #
+      ########################
 
-        
-        # Standard error. Known variance, sample with replacement.
-        def standard_error_ksd_wr(s, sam, pop)
-            s.quo(Math::sqrt(sam)) * Math::sqrt((pop-1).quo(pop))
-        end
-        
-        # Standard error of the mean. Known variance, sample w/o replacement
-        def standard_error_ksd_wor(s,sam,pop)
-            s.quo(Math::sqrt(sam)) * Math::sqrt(qf(sam,pop)) 
-        end
-        
-        alias_method :standard_error_esd_wr, :standard_error_ksd_wr
-        
-        # Standard error of the mean. 
-        # Estimated variance, without replacement
-        # Cochran (1972) p.47
-        def standard_error_esd_wor(s,sam,pop)
-            s.quo(Math::sqrt(sam)) * Math::sqrt(qf(sam,pop))
-        end
-        
-        alias_method :standard_error, :standard_error_esd_wor
-        alias_method :se, :standard_error_esd_wor
+      
+      # Standard error. Known variance, sample with replacement.
+      def standard_error_ksd_wr(s, sam, pop)
+          s.quo(Math::sqrt(sam)) * Math::sqrt((pop-1).quo(pop))
+      end
+      
+      # Standard error of the mean. Known variance, sample w/o replacement
+      def standard_error_ksd_wor(s,sam,pop)
+          s.quo(Math::sqrt(sam)) * Math::sqrt(qf(sam,pop)) 
+      end
+      
+      alias_method :standard_error_esd_wr, :standard_error_ksd_wr
+      
+      # Standard error of the mean. 
+      # Estimated variance, without replacement
+      # Cochran (1972) p.47
+      def standard_error_esd_wor(s,sam,pop)
+          s.quo(Math::sqrt(sam)) * Math::sqrt(qf(sam,pop))
+      end
+      
+      alias_method :standard_error, :standard_error_esd_wor
+      alias_method :se, :standard_error_esd_wor
 
-        # Standard error of total estimation
-        
-        def standard_error_total(s,sam,pop)
-            pop*se(s,sam,pop)
-        end
+      # Standard error of total estimation
+      
+      def standard_error_total(s,sam,pop)
+          pop*se(s,sam,pop)
+      end
 
-        # Confidence Interval using T-Student
-        # Use with n < 60
-        def mean_confidence_interval_t(mean,s,n_sample,n_population,margin=0.95)
-            t=Distribution::T.p_value(1-((1-margin) / 2),n_sample-1)
-            mean_confidence_interval(mean,s,n_sample,n_population,t)
-        end
-        # Confidente Interval using Z
-        # Use with n > 60
-        def mean_confidence_interval_z(mean,s,n_sample,n_population,margin=0.95)
-            z=Distribution::Normal.p_value(1-((1-margin) / 2))
-            mean_confidence_interval(mean,s,n_sample,n_population, z)
-        end
-        # Confidente interval using X.
-        #
-        # Better use mean_confidence_interval_z or mean_confidence_interval_t
-        def mean_confidence_interval(mean,s,n_sample,n_population,x)
-            range=x*se(s,n_sample,n_population)
-            [mean-range,mean+range]
-        end
+      # Confidence Interval using T-Student
+      # Use with n < 60
+      def mean_confidence_interval_t(mean,s,n_sample,n_population,margin=0.95)
+          t=Distribution::T.p_value(1-((1-margin) / 2),n_sample-1)
+          mean_confidence_interval(mean,s,n_sample,n_population,t)
+      end
+      # Confidente Interval using Z
+      # Use with n > 60
+      def mean_confidence_interval_z(mean,s,n_sample,n_population,margin=0.95)
+          z=Distribution::Normal.p_value(1-((1-margin) / 2))
+          mean_confidence_interval(mean,s,n_sample,n_population, z)
+      end
+      # Confidente interval using X.
+      #
+      # Better use mean_confidence_interval_z or mean_confidence_interval_t
+      def mean_confidence_interval(mean,s,n_sample,n_population,x)
+          range=x*se(s,n_sample,n_population)
+          [mean-range,mean+range]
+      end
 		end
 	end
-        
-    end
+end