Data processing



News articles
1.https://kyivindependent.com/ukrainians-trust-in-zelensky-decreasing-but-still-high-survey-shows/





•



Data in acadamic articles and books



Real Data



Data linked to research articles
•Processed data
•Edited data
•Original variables not used for article are deleted
•Indication of source of data
•
•data.mendeley.com
•https://dataverse.harvard.edu/
•
•

•Vykopirovat vietnam



Typical scenarios
•Excel  (xls, xlsx)
•SPSS (sav), Stata (dta), R (rdata)
•CSV, TXT
•XML
•JSON

Examples



SPSS
SPSS is available at the MU website (see next page)
SPSS to excel online conversion: https://secure.ncounter.de/spssconverter
https://www.afrobarometer.org/survey-resource/merged-round-8-data-34-countries-2022/

•https://inet.muni.cz/?idr=ict.soft&pk=u&x=0&y=500



CSV
Our example of csv data:
https://data.europarl.europa.eu/en/datasets/texts-adopted-by-the-european-parliament-year2023/26




•
>
TT
This is important for tables containing text in other language than English

Split column into more columns



Cleaning (ctr+h)
•


Sorting



This box has to be marked to see variable names in „sort by“
Here you can add one more box for choosing sorting variable


Computation
Grouping variable: for what categories we want to make the computation
Way of computation: what kind of number we want (sum, count, mean, min, max, SD)
Sum: the ammount which you have to pay in shop is sum of prices of pieces in your shopingcart
Count: number of pieces in your shopingcart
Mean: average price of pieces (e.g. 20, 30, 50, 100 mean is (20+30+50+100/4 = 50)
SD: mean deviation of values around the mean (50-20) + (50-30) + (50-50) +(100-50)/4 = 25
      (real calculation works with squareroot of exposed numbers
What you want to count

Result
•It is not over yet


•Copy the results to new sheet and insert them as values
•Click with right button of mouse
•Select only rows wit result
•Add new empty column
•Use text to column wit „space“ as delimitor
•Sort data by new column
•Delete everything without word count (in new column)

Compute percentages
•=C2/$C$70*100
•
•$ allows to calculate values for whole column, but the value in denominator remains the sam