差分

ナビゲーションに移動 検索に移動
Suppose we have such a dataset in STATA.==Abbreviations of commands=={|class="wikitable"|-!table[[file:STATAsample|''(no abbv.jpg]])''|-==Differences between table, !tabulate, |'''ta'''<br>'''tab'''|-!tabstat, |''(no abbv.)''|-!summarize==|'''su'''|}
==Differences between '''table''', '''tabulate''', '''tabstat''', '''summarize'''==
{|class="wikitable"
|-
!table
|<pre>table v1</pre>
create a one-way table<br> of ''v1'' <br> with '''simple''' statisticsfrequency
|<pre>table v1 v2</pre>
create a two-way table<br> of ''v1'' <br> in row&dagger; and ''v2'' in column&dagger;
|<pre>,statistics( )</pre>
|-
!tabulate
|<pre>tabulate v1</pre>
create a one-way table<br> of ''v1'' <br> with '''detailed''' statisticsfrequency
|<pre>tabulate v1 v2</pre>
create a two-way table<br> with ''v1'' <br> in row&dagger; and ''v2'' in column&dagger;
|<pre>,chi2</pre>
Pearson's chi-squared test; <nowiki>*</nowiki>only for two-way
!tabstat
|<pre>tabstat v1</pre>
create a one-way table of ''v1''<br> with '''detailed ''' statistics
|''*no two- or multiple-way table''
|<pre>,statistics( )</pre>
!summarize
|<pre>summarize v1</pre>
'''detailed ''' statistics of ''v1''
|''*no two- or multiple-way summary''
|<pre>,detail</pre>
&dagger; row = transverse direction, column = longitudinal direction
==Sample data=Abbreviations =Suppose we have such a dataset in STATA. [[file:STATAsample.jpg]] Where,{|!id|''discrete''|:Identification number|-!sex|''binary''|:Male=0, Female=1|-!data1|''continuous''|:Results of a certain test|-!factorA, B, C|''binary''|:Negative=0, Positive=1|-!SES|''categorical''|:Categories of Socio-Economic Status, divided into four|-!disease|''binary''|:Free from a certain disease=0, Having the disease=1|} ==One-way=====Summary of ''sex'', a binary variable==={|class="wikitable"|-!table|'''table sex'''[[file:table_sex.jpg]]|rowspan="2"|Both reports frequency<br> but '''tabulate''' is more detailed|-!tabulate|'''tabulate sex'''[[file:tabulate_sex.jpg]]|-!tabstat|'''tabstat sex'''[[file:Tabstat_sex.jpg]]|rowspan="2"|Both reports mean<br> but '''summarize''' is more detailed|-!summarize|'''summarize sex'''[[file:summarize_sex.jpg]]|} ===Summary of ''data1'', a continuous variable==={|class="wikitable"|-!table|'''table data1'''[[file:table_data1.jpg]]|rowspan="2"|Both reports frequency of each value,<br> which does not make sense|-!tabulate|'''tabulate data1'''[[file:tabulate_data1.jpg]]|-!tabstat|'''tabstat data1'''[[file:tabstat_data1.jpg]]|rowspan="2"|Both reports mean<br> but '''summarize''' is more detailed|-!summarize|'''summarize data1'''[[file:summarize_data1.jpg]]|} ===Summary of ''SES'', a categorical variable==={|class="wikitable"|-!table|'''table SES'''[[file:table_SES.jpg]]|rowspan="2"|Both reports frequency<br> but '''tabulate''' is more detailed|-!tabulate|'''tabulate SES'''[[file:tabulate_SES.jpg]]|-!tabstat|'''tabstat SES'''[[file:tabstat_SES.jpg]]|rowspan="2"|Both reports mean<br> but '''summarize''' is more detailed|-!summarize|'''summarize SES'''[[file:summarize_SES.jpg]]|} ==One-way, multiple=={|class="wikitable"|-!table|rowspan="2" colspan="2"|''*Both do not create one-way multiple table''|-!tabulate|-!tabstat|'''tabstat sex data1 SES'''[[file:tabstat_sex_data1_SES.jpg]]|Reports mean in row (transverse) direction|-!summarize|'''summarize sex data1 SES'''[[file:summarize_sex_data1_SES.jpg]]|Reports more details in column (longitudinal) direction|} ==Two-way=====Summary of ''factorA'' based on ''sex''==={|class="wikitable"|-!table|'''table sex factorA'''[[file:table_sex_factorA.jpg]]|rowspan="2"|Both creates the same table<br> but '''tabulate''' is better visualized|-!tabulate|'''tabulate sex factorA'''[[file:tabulate_sex_factorA.jpg]]|-!tabstat|'''tabstat factorA, by(sex)'''[[file:tabstat_factorA_by(sex).jpg]]|rowspan="2"|Both reports mean<br> but '''summarize''' is more detailed;<br>needs '''bysort''' option before the command|-!summarize|'''bysort sex: summarize factorA'''[[file:summarize_factorA_bysort_sex.jpg]]|} ===Summary of ''sex'' based on ''factorA''==={|class="wikitable"|-!table|'''table factorA sex'''[[file:table_factorA_sex.jpg]]|rowspan="2"|Both creates the same table<br> but '''tabulate''' is better visualized|-!tabulate|'''tabulate factorA sex'''[[file:tabulate_factorA_sex.jpg]]|-!tabstat|'''tabstat sex, by(factorA)'''[[file:tabstat_sex_by(factorA).jpg]]|rowspan="2"|Both reports mean<br> but '''summarize''' is more detailed;<br>needs '''bysort''' option before the command|-!summarize|'''bysort factorA: summarize sex'''[[file:summarize_sex_bysort_factorA.jpg]]|} ===Summary of commands''data1'' based on ''disease''===
{|class="wikitable"
|-
!table
|rowspan="2" colspan="2"|''(no abbv.)*Both do not create a meaningful table for continuous variable''
|-
!tabulate
|ta<br>tab
|-
!tabstat
|'''tabstat data, by(no abbvdisease)'''[[file:tabstat_data1_by(disease).)jpg]]|rowspan="2"|Both reports mean<br> but '''summarize''' is more detailed;<br>needs '''bysort'''option before the command
|-
!summarize
|su'''bysort disease: summarize data1'''[[file:summarize_data1_bysort_disease.jpg]]|} ==Two-way with proportions=====Summary of ''factorA'' based on ''sex'' with proportions==={|class="wikitable"|-!rowspan="3"|table|'''table sex factorA, statistic(percent)'''[[file:table_sex_factorA_statistic(percent).jpg]]|This calculates proportions of cells compared to the whole<br> without showing raw values|-|'''table sex factorA, statistic(percent, across(sex))'''[[file:table_sex_factorA_statistic(percent_across(sex)).jpg]]|This calculates proportions in column (longitudinal) directions<br> without showing raw values |-|'''tale sex factorA, statistic(percent, across(factorA))'''[[file:table_sex_factorA_statistic(percent_across(factorA)).jpg]]|This calculates proportions in row (transverse) directions<br> without showing raw values |-!rowspan="2"|tabulate|'''tabulate sex factorA, column'''[[file:tabulate_sex_factorA_column.jpg]]|This calculates proportions in column (longitudinal) directions|-|'''tabulate sex factorA, row'''[[file:tabulate_sex_factorA_row.jpg]]|This calculates proportions in row (transverse) directions|} ==Two-way, multiple=====Summary of ''factorA'', ''factorB'', ''factorC'' based on ''disease''==={|class="wikitable"|-!rowspan="3"|tabstat|'''tabstat factorA factorB factorC, by(disease)'''[[File:tabstat_factorABC_by(disease).jpg]]||-|'''tabstat factorA factorB factorC, by(disease) statistic(sum)'''[[File:tabstat_factorABC_by(disease)_statistic(sum).jpg]]|factorA,B,C are binary variables so summations of values provide the positivities of factorA,B,C|-|'''tabstat factorA factorB factorC, by(disease) statistic(n)'''[[File:tabstat_factorABC_by(disease)_statistic(n).jpg]]|''statistic(n)'' (''statistic(count)'' is the same) only counts observations with real values, which only tell non-missing observations|} ===Summary of ''factorA'', ''factorB'', ''factorC'' based on ''SES''==={|class="wikitable"|-!rowspan="3"|tabstat|'''tabstat factorA factorB factorC, by(SES)'''[[File:tabstat_factorABC_by(SES).jpg]]||-|'''tabstat factorA factorB factorC, by(SES) statistic(sum)'''[[File:tabstat_factorABC_by(SES)_statistic(sum).jpg]]|factorA,B,C are binary variables so summations of values provide the positivities of factorA,B,C|-|'''tabstat factorA factorB factorC, by(SES) statistic(n)'''[[File:tabstat_factorABC_by(SES)_statistic(n).jpg]]|''statistic(n)'' (''statistic(count)'' is the same) only counts observations with real values, which only tell non-missing observations|} ==Two-way of binary/categorical plus summary of continuous=====Summary of ''data1'' based on ''disease'' and ''SES''==={|class="wikitable"|-!tabulate|'''tabulate disease SES, summarize(data1)'''[[File:tabulate_disease_SES_summarize(data1).jpg]]|This tells means, SDs and frequencies of a continuous variable divided in two-way of binary/categorical variables
|}

案内メニュー