Merge pull request #4 from daelsaid/mturk

mturk_extra_measures

post_scoring_compiled_csv/vmreact_extra_measures_combined.py

@@ -0,0 +1,34 @@
+import pandas as pd
+import os
+import sys
+import csv
+from glob import glob
+import numpy as np
+
+data_dir='/Users/lillyel-said/Desktop/vmreact/vmreact/2_vmreact/'
+compiled='/Users/lillyel-said/Desktop/vmreact/vmreact/2_vmreact/mturk_vmreact_complete_compilation_initiation.csv'
+av_typing='/Users/lillyel-said/Desktop/vmreact/vmreact/2_vmreact/typing_test_averages.csv'
+trials=['trial1','trial2','trial3','trial4','trial5','listb','trial6','trial7']
+cols=['values.response_latency', 'expressions.trial_recall_word_latency',
+      'values.recall_firstcharlatency', 'values.recall_lastcharlatency']
+column_titles=['subjid','date']
+
+incorrect=pd.read_csv(os.path.join(data_dir,'incorrect_response.csv'),dtype=str)
+vmreact_df=pd.read_csv(compiled,dtype='str')
+new_compiled=pd.DataFrame(data=df2,dtype=str)
+avg_typing_df=pd.read_csv(av_typing,dtype='str')
+
+vmreact_df['unique_identifier']=vmreact_df['subject']+'_'+vmreact_df['date']
+df2=vmreact_df.merge(avg_typing_df,left_on='unique_identifier',right_on='unique_identifier',how='outer')
+
+
+new_compiled["total_average_repeats"]=new_compiled.loc[:,'listb_#_repeats':'trial7_#_repeats'].astype(float).mean(axis=1)
+new_compiled["total_incorrect"]=new_compiled.loc[:,'listb':'trial7'].astype(float).subtract(15,axis=0)
+new_compiled[['listb_errors','trial1_errors','trial2_errors','trial3_errors','trial4_errors','trial5_errors','trial6_errors','trial7_errors']]=new_compiled.loc[:,'listb':'trial7'].astype(float).subtract(15,axis=0).abs()
+
+# for x,y in incorrect.groupby(['subj_id','trial']):
+#     print x,y.score.value_counts().T
+
+new_compiled.to_csv('updated_mturk_vmreact_complete_compilation_initiation.csv')
+
+

post_scoring_compiled_csv/vmreact_gen_normed_tables.py

@@ -0,0 +1,40 @@
+import pandas as pd
+import os
+import sys
+from glob import glob
+import numpy as np
+
+
+data_dir='/Users/lillyel-said/Desktop/vmreact/vmreact/2_vmreact/'
+
+cols=['date','subject', 'trial1','trial2','trial3','trial4','trial5','listb','trial6','trial7','total_learning','corrected_total_learning','learning_rate','proactive_interference','retroactive_interference','forgetting_and_retention']
+trials=['trial1', 'trial2', 'trial3', 'trial4', 'trial5', 'listb', 'trial6', 'trial7']
+
+
+for compiled_file in glob(os.path.join(data_dir,'mturk_vmreact_complete_compilation_initiation.csv')):
+    vmreact_compiled=pd.read_csv(compiled_file,dtype=str,index_col=['gender','age_range'])
+    bysubj=pd.read_csv(compiled_file,dtype='str')
+    test_df=vmreact_compiled.loc[:,'list_type':'trial7_values.recall_lastcharlatency']
+    for t in trials:
+        for subj,subj_df in vmreact_compiled.groupby(level=[0,1]):
+            if len(subj_df) >3:
+                try:
+                    response_latency=subj[0], subj[1],t+'_values.response_latency', round(subj_df[t+'_values.response_latency'].astype(float).mean(axis=0),4),round(subj_df[t+'_values.response_latency'].astype(float).std(axis=0),4),subj_df[t+'_values.response_latency'].count()
+                    initiation=subj[0],subj[1], t+'_values.recall_firstcharlatency', round(subj_df[t+'_values.recall_firstcharlatency'].astype(float).mean(axis=0),4),round(subj_df[t+'_recency'].astype(float).std(axis=0),4),subj_df[t+'_recency'].count()
+                    repeats=subj[0],subj[1],t+'_#_repeats',round(subj_df[t+'_#_repeats'].astype(float).mean(axis=0),4),round(subj_df[t+'_#_repeats'].astype(float).std(axis=0),4),subj_df[t+'_#_repeats'].count()
+                    trials=subj[0],subj[1],t,round(subj_df[t].astype(float).mean(axis=0),4),round(subj_df[t].astype(float).std(axis=0),4),subj_df[t].count()
+                    primacy=subj[0],subj[1],t+'_primacy',round(subj_df[t+'_primacy'].astype(float).mean(axis=0),4),round(subj_df[t+'_primacy'].astype(float).std(axis=0),4),subj_df[t+'_primacy'].count()
+                    recency=subj[0],subj[1],t+'_recency',round(subj_df[t+'_recency'].astype(float).mean(axis=0),4),round(subj_df[t+'_recency'].astype(float).std(axis=0),4),subj_df[t+'_recency'].count()
+                    composite=subj_df.loc[:,'total_learning':'forgetting_and_retention'].astype(float)
+                    composite_vals=composite.mean(axis=0), composite.std(axis=0),composite.count()
+                    comp=composite.mean(axis=0), composite.std(axis=0),composite.count()
+                    # # print comp[2].T
+                    # print repeats
+                    # print response_latency
+                    # print initiation
+                    # print trials
+                    # print primacy
+                    # print recency
+                except:
+                    continue
+                    #firstcharaverages

post_scoring_compiled_csv/vmreact_typing_latency_by_subject.py

@@ -0,0 +1,13 @@
+import pandas as pd
+from glob import glob
+import os
+import shutil
+import sys
+
+
+final=[]
+
+for x in glob(os.path.join(data_dir,'filtered_typing_test.csv')):
+    df=pd.read_csv(x, dtype='str')
+    for i, val in df.groupby(['subject','date']):
+        print i[0]+'_'+i[1],i[0],i[1],val['latency'].astype(int).mean()