positivity

#! /usr/bin/env ruby
require "twitter"
require 'csv'
require 'regexp'

# Twitter setup was here...

#Gets the last 10 tweets from a given user from the twitter API
screen_name = String.new ARGV[0]

#for each tweet 
tweet = client.user_timeline(screen_name, count: 10).each do |tweet|

    #Look through each row in the CSV file to see if any words in the tweet match the words in the 2nd element of the csv array(the word)
    csv = CSV.foreach("csvratings.csv", {:headers=>true}).each do |row|)

	#For each line if the word matches words in the tweet then create a new array and put the valence mean value (3rd array element) in it.
	if row[1].match(tweet)do
		valence = array.new 
		valance.push(row[2])

		#Find the average of the valence array to be ordered later
		len = valence.length
		total = valence.inject(:+)
        average = total.to_f / len

    #print out the tweet and their average valence number
    puts tweet + average

        #Alternative ending idea...
        #average valence number needs to be ordered, not sure how to get these to correspond with the tweets... 
        all_averages = array.new
        all_averages.push(average + tweet)
        all_averages.sort

        puts tweet + all_averages


#Notes:
#For the tweets each count search through the CSV loop,
#if any words match then assign them with the coresponding number in vmean[2] of the CSV array
#Find the valence avergage for the words in each line 
#Assign them a value from 1 - 10 and print in that order

OK, good start, a couple of things that might help you crack this one:

Creating and adding to arrays

This is a perfectly valid way of creating an array:

my_array = Array.new

But unless you've got a good reason for doing so (you want to set an initial size and default value), we usually just do this:

my_array = []

Adding things to an array with my_array.push(thing_to_add) is perfectly idiomatic Ruby, my personal preferences is to use the double arrows instead:

my_array << 'cheese'
my_array << 'on'
my_array << 'toast'

# at this stage, my_array is ['cheese', 'on', 'toast']

my_array << 'is' << 'probably' << 'not' << 'good' << 'you'

# by now, my_array is ['cheese', 'on', 'toast', 'is', 'probably', 'not', 'good', 'for', 'you']

Indentation

You're almost definitely going to get confused about which do..end block you're in if you're indenting as you go!

Other notes

• I don't thing you need the require 'regexp' directive, could be wrong.
• if statements don't require an enclosing do for the code that they branch on, but they do require a finishing end.
• Don't worry about the order of the tweets, they just need the numeric rating next to them.