An order line arrives from a messy export: the unit price reads $129.00, the quantity reads 2, and a 15% coupon was applied. A human reads that and calculates the line revenue in their head. Python cannot, yet: to Python, "$129.00" is a piece of text, no more a number than the word "skillet". Knowing what kind of value you are holding is the first skill of data work.
A value is a piece of data: 2, 129.0, "Cast Iron Skillet", True. A variable is a name you attach to a value with = so you can use it later.
product = \"Cast Iron Skillet 26cm\"\nunit_price = 39\nquantity = 2\nline_total = unit_price * quantity\nprint(product, line_total)Read line_total = unit_price * quantity from right to left: Python computes the right side first, then points the name on the left at the result. Variable names use lowercase words joined by underscores (unit_price, not UnitPrice or unitprice); that convention is called snake_case and every Python codebase you will read follows it.
The type decides what you can do with a value. type() reveals it:
print(type(2)) # int: whole numbers\nprint(type(129.0)) # float: numbers with a decimal point\nprint(type(\"129.00\")) # str: text, in single or double quotes\nprint(type(True)) # bool: True or False\nprint(type(None)) # NoneType: \"no value here\"Notice that "129.00" is a str. Quotes make text, even when the characters inside look like digits. That single fact explains a whole family of data bugs: a column that looks numeric but was read as text sorts "100" before "20" and refuses to sum.
Numbers support the operators you expect, plus three that matter in analysis:
\nprint(10 / 4) # 2.5 true division, always a float\nprint(10 // 4) # 2 floor division: whole boxes of 4\nprint(10 % 4) # 2 remainder: items left over\nprint(2 ** 3) # 8 power\nprint(0.1 + 0.2) # 0.30000000000000004The last line is not a Python bug. Computers store decimals in binary, so most decimal fractions are stored as a tiny approximation. For money you round at the end with round(value, 2), and you never compare two floats with == after a calculation unless you rounded both. Section 4 returns to this when totals do not quite match.
One more surprise hides in round() itself. Python rounds exact halves to the nearest even digit, so round(0.5) is 0 and round(2.5) is 2, and a value like 2.675 is stored as slightly less than it looks, so round(2.675, 2) gives 2.67. For analysis and reporting that never matters. For invoices, where every cent must follow an accounting rule, Python’s decimal module exists; you will not need it in this course, but now you know where to look when a finance colleague asks.
Strings come with methods: functions attached to the value, called with a dot. These are the ones you will use constantly:
\nraw = \" united KINGDOM \"\nprint(raw.strip()) # remove spaces at both ends\nprint(raw.strip().title()) # \"United Kingdom\"\nprint(raw.upper()) # all capitals\nprint(\"$129.00\".replace(\"$\", \"\")) # swap one substring for another\nprint(len(\"Cartwheel\")) # number of characters: 9Every one of these returns a new string. Strings are immutable: raw.strip() does not change raw. If you want to keep the cleaned version, assign it: country = raw.strip().title().
int(), float() and str() convert values, but only when the conversion makes sense:
quantity = int(\"2\")\nprice = float(\"129.00\")\nlabel = \"Qty: \" + str(quantity)\nprint(quantity * price, label)float("$129.00") fails with ValueError: could not convert string to float: '$129.00'. The fix is to clean first, then convert. You chain methods left to right, each one working on the result of the previous one:
price_text = \" $129.00\"\nprice = float(price_text.strip().replace(\"$\", \"\"))\ndiscount = 0.15\nline_revenue = round(2 * price * (1 - discount), 2)\nprint(price, line_revenue)That is the revenue formula Cartwheel uses for every order line: quantity times unit price times one minus the discount. You will apply it to thousands of rows at once in Section 2; here you see it for one.
\n\nComparisons produce True or False, and these answer yes/no questions about your data:
line_revenue = 219.3\nprint(line_revenue >= 200) # True\nprint(line_revenue == 219.3) # True: equal values\nprint(\"web\" != \"mobile_app\") # True: not equalOne = assigns; two == compares. Mixing them up is the most common typo for new programmers, and Python helps by rejecting if x = 3: as a syntax error.
None is a separate value that means “nothing here”. A missing coupon code or an unknown satisfaction score would be None in plain Python. pandas uses its own markers for missing data, and Section 3 is about handling them carefully.
The exercise below turns one messy export row into real numbers. Next, you store many values at once in lists and dictionaries.
\n","toc":[{"id":"sec-1","text":"Values and variables","level":2},{"id":"sec-2","text":"Every value has a type","level":2},{"id":"sec-3","text":"Arithmetic","level":3},{"id":"sec-4","text":"Text is a sequence of characters","level":2},{"id":"sec-5","text":"Converting between types","level":2},{"id":"sec-6","text":"Booleans and comparisons","level":2}],"takeaways":["A variable is a name pointing at a value;= assigns, == compares.","Every value has a type, and type() tells you which one; most bugs in data work are a value having the type you did not expect.","/ always returns a float, // drops the remainder, and % gives the remainder.","String methods such as strip() and replace() return a new string; they never change the original.","Convert text to numbers with int() and float() only after removing symbols like $ and spaces."],"further":[{"title":"An Informal Introduction to Python (numbers and text)","url":"https://docs.python.org/3/tutorial/introduction.html"},{"title":"Built-in Types - String Methods","url":"https://docs.python.org/3/library/stdtypes.html#string-methods"},{"title":"Floating-Point Arithmetic: Issues and Limitations","url":"https://docs.python.org/3/tutorial/floatingpoint.html"}],"quiz":[{"q":"What does this print?
\nprice = \"39\"\nprint(price * 2)78","why":"That would need int(price) * 2. As written, price is text."},{"text":"3939","why":"Correct. Multiplying a string by an integer repeats it. The value came in as text, so the maths never happened."},{"text":"A TypeError","why":"str * int is allowed in Python (it repeats the string). The error happens with str + int."}],"answer":1},{"q":"Which line turns "$1,250.00" into the float 1250.0?
float("$1,250.00")","why":"float() accepts digits, a decimal point, a sign, an exponent and surrounding spaces, but not $ or a thousands comma, so this raises ValueError."},{"text":"int("$1,250.00".replace("$", ""))","why":"The comma is still there, and int() would reject the .00 anyway."},{"text":"float("$1,250.00".replace("$", "").replace(",", ""))","why":"Correct. Remove every symbol first, then convert the clean digits."},{"text":"str("$1,250.00").strip()","why":"strip() only removes whitespace at the ends, and the result is still a string."}],"answer":2},{"q":"What is the value of 10 / 4 and 10 // 4?
2.5 and 2","why":"Correct. True division always returns a float; floor division drops the fractional part."},{"text":"2 and 2.5","why":"It is the other way round. // is the operator that rounds down."},{"text":"2.5 and 2.5","why":"// is floor division and never keeps the fractional part."}],"answer":0},{"q":"A customer’s coupon code is unknown. Which value best represents that in Python?
","options":[{"text":"The empty string""","why":"An empty string is a real value (text with no characters). It can mean “no coupon” or “unknown”, which is exactly the ambiguity you want to avoid."},{"text":"The number 0","why":"Zero is a real number and would be counted in sums and averages as if it were data."},{"text":"None","why":"Correct. None is Python’s built-in marker for “no value here”. pandas has its own missing markers, which you meet in Section 3."}],"answer":2}],"exercise":{"kind":"python","files":[],"starter":"# One row from Cartwheel's messy export, exactly as it arrived\nprice_text = \" $129.00\"\nquantity_text = \"2\"\ndiscount = 0.15\n\n# 1. Turn price_text into a float\nprice = None\n\n# 2. Turn quantity_text into an int\nquantity = None\n\n# 3. Line revenue = quantity * price * (1 - discount), rounded to 2 decimals\nline_revenue = None\n\n# 4. True when line_revenue is 200 or more\nis_large = None\n","solution":"price_text = \" $129.00\"\nquantity_text = \"2\"\ndiscount = 0.15\n\nprice = float(price_text.strip().replace(\"$\", \"\"))\nquantity = int(quantity_text)\nline_revenue = round(quantity * price * (1 - discount), 2)\nis_large = line_revenue >= 200\n","checks":[{"id":"price-float","test":"type(price) is float and price == 129.0"},{"id":"quantity-int","test":"type(quantity) is int and quantity == 2"},{"id":"revenue","test":"isinstance(line_revenue, float) and line_revenue == 219.3"},{"id":"is-large","test":"is_large is True"}],"title":"Turn a messy row into numbers","prompt":"A row from the messy export gives you price_text = " $129.00", quantity_text = "2" and discount = 0.15. Create price (a float), quantity (an int), line_revenue (quantity × price × (1 − discount), rounded to 2 decimals) and is_large (a boolean: True when line_revenue is 200 or more). Calculate from the variables; don’t type the answers in by hand.
float() and int() convert text, but only clean text: digits and a decimal point.","Remove the space with .strip() and the dollar sign with .replace("$", "") before calling float().","line_revenue = round(quantity * price * (1 - discount), 2) and is_large = line_revenue >= 200."],"explanation":"price_text.strip().replace("$", "") produces "129.00", which float() accepts. int("2") gives the integer 2. The revenue formula multiplies three numbers; without round() you would get 219.29999999999998 because of how floats are stored. A comparison like line_revenue >= 200 already evaluates to True or False, so you assign it directly; there is no need for an if.
price is the float 129.0","quantity-int":"quantity is the integer 2","revenue":"line_revenue is 219.3","is-large":"is_large is True"}},"words":720,"readMinutes":3}