Addendum

## ETL PROCESS (extract, transform, load)							
							
loaded bfro._reports.json file into tableau to explore. 							
took note of col's to join with other tables.							
exported as .csv file							
loaded bfro.json.csv files along with locations and geocoded csv s into sheets							
extracted dates on all three							
noticed time stamp was incorrect all at 12:00 deleted col, kept day month year							
							
geocoded csv cleaning							
deleted empty rows							
removed duplicates							
split report_title into 2 col's, report_number & report_description.  Deleted report_description as it was in another table.							
deleted precip_intensity col not needed							
noticed only date included was a col called day and the rows were sporadic with numbered days only, changed format to string							
I kept the col 'day' because when filtered AZ on all three sheets, report_number, class and day were consistent and matching.							
deleted col's I didn't need: temperature_hightemperature_midtemperature_lowdew_pointhumidity							
created new columns to calculate mean, median, mode, std. deviation, variance and quantiles of the fractional moon light  per each recorded sighting							
copied values to new col's and delected the calculated col's							
							
Created new tab called quartiles							
copied values over to new tab for safe keeping							
							
bfro_locations cleaning							
found and replaced all characters after dates in timestamp, as the time was incorrect at 12:00:00 for all incidences.							
extracted day, month, year from what was left							
copied and pasted values only into new col's, deleted extracted col's							
split report_title into 2 col's, report_number & kept report_description				<--- to match on geocoded table			
							
bfro_reports cleaning							
date was 1974-09-20, and timestamp 1974-09-20T12:00:00Z							
extracted date to day month year, deleted timestamp as it didn't give time							
changed dates to strings so can match to geocoded table							
cleaned up col names for clarity, changed year to string							
deleted col's nearest town and nearest rd.  I have latitude, longitude, county and country							
							
moon calendar cleaning		<--- not sure I am going to use this table					
I extracted the date and time from time stamp to their own col's							
copied the values to new col's and deleted the extracted cols							
I then deleted all rows before the dat4e of 1921 and after the date of 2021 to match the year data I have for bigfoot sightings							
							
I changed the names on all tabs							
reports							
locations							
geocode							
full_moon_cal							
and created a new one							
moon_days	<-- I determined each day of the moonth by the fractional representation of the moon shine off of the geocode table.						
Looking again at the columns on different tables I've confirmed each table has matching report number id's to match on. 							
I marked the col's 'season' on two tables, i can get rid of one							
Fields to join on in tablau will be record_number, day, and class							
some columns were duplicated on geocode, locations and reports. I deleted as seemed appropriate.