VCE Data Analytics - U3O1 - Data Analytics Hangman

The 141 key terms

1Interpreting a brief

Reading a teacher's brief to find the data, calculations and outputs needed, the constraints that limit the solution, and what is out of scope.

2The solution must

Starts a functional requirement: one thing a digital solution must do. It says what the solution does, not how, and covers one action. Example: The solution must show the score.

3Non-functional requirement

A quality describing how well a system performs, such as speed, security, or reliability, rather than what it does

4measurable non-functional requirement

A quality of a solution, such as speed, security or usability, written with a testable target, like a page loads in under 2 seconds, so you can check whether it is met.

5constraint

A limit on how a solution can be made, such as time, cost, skills, devices or software. A constraint is not a feature; it limits what you can build.

6scope

The boundary of what a solution will and will not include. Features inside the boundary are in scope. Features left out, or saved for later, are out of scope.

7requirements list

A list that sorts what a solution must do (requirements) from the limits it must work within (constraints), for one problem. Each item should be clear and testable.

8string

Text made of a sequence of characters (single letters, digits or symbols), written in quotes, like 'Sam' or '0412345678'. Phone numbers are strings.

9integer

A whole number with no decimal part, positive or negative, used for counts such as the number of students in a class (e.g. 25, 0 or -3).

10float

A floating point number is a number with a decimal part, such as 1.75 or 19.95. It is used for measurements like height and for money, unlike an integer, which has no decimals.

11Date/Time

A data type that stores a calendar date and/or time of day, such as 25/03/2025 or 14:30, so values can be sorted in order and used to calculate the duration between two dates.

12Boolean

A data type with only two possible values: true or false (yes or no). Used for yes/no fields, TRUE and FALSE in a spreadsheet, and the result of a comparison such as 5>3.

13data type

Pick the data type by what you will do with the value: calculate (number), sort by date (date/time), test true or false (Boolean). Postcodes are text, not numbers, as you never do maths on them.

14relational database

A database that stores data in separate tables of fields (columns) and records (rows), with tables linked by shared values such as keys.

15database table

A database table stores data in columns called fields and rows called records. A single-table database keeps all its data in this one table.

17VARCHAR(30)

Field size sets the maximum length or storage a value may use. VARCHAR(30) allows up to 30 characters; longer values are rejected or cut. Smaller sizes save storage.

18VARCHAR(50)

A column's data type and size set in CREATE TABLE. VARCHAR(50) holds text up to 50 characters. Other types: INT, DECIMAL(8,2), DATE, BOOLEAN. They control what a column stores.

19PRIMARY KEY

A field whose value is unique for every record in a table, such as a student ID number, so each record can be identified exactly. Names make poor keys because they can repeat.

20FOREIGN KEY

A field in one table that holds the primary key value of a record in another table, linking the two tables together so related data is stored once and not repeated.

21one-to-many relationship

One record in a primary table matches many records in a related table. The foreign key sits on the many side, e.g. one class has many students, so class details are not repeated.

22CREATE TABLE

SQL command that builds a new table by naming its fields and their data types, such as a PRIMARY KEY. INSERT INTO ... VALUES then adds one record to that table.

23SELECT name, age FROM students;

SELECT lists the fields to show and FROM names the table. This returns the name and age columns for every record in students. SELECT * shows all columns. End with a semicolon.

24WHERE

An SQL clause that keeps only the rows meeting a condition, such as WHERE age >= 15 or WHERE town = 'Perth'. Text goes in quotes, and AND or OR can combine conditions.

25AND, OR, NOT

Combine WHERE conditions: AND needs both true, OR needs at least one true, NOT reverses a condition. Brackets set the order and decide which rows are returned.

26ORDER BY

Sorts the rows a SELECT query returns by a chosen field. It sorts A to Z or lowest to highest by default (ASC), or the reverse if you add DESC after the field name.

27INNER JOIN t2 ON t1.key = t2.key

Combines rows from two tables where a foreign key matches a primary key, so related data shows in one result. Use table.field names to say which table each field belongs to.

28Export query results as CSV

Save the rows a query returns as a CSV file with a header row of field names. Import it into a spreadsheet, then check the columns came across correctly before further analysis.

29data dictionary

A table listing each data element with its name, data type, size or format, description and validation rule, so everyone uses the data the same way when coding and testing.

30studentAge, integer, 2 digits, age in years, 15, 11 to 19

One data dictionary row for a data item: name, data type, format or size, description, example value and validation rule. Written from a requirement or IPO chart.

31IPO chart

A three-column table listing a solution's inputs (data entered), the processes applied to that data, and the outputs it produces, such as a total or report.

33Query design

A plan for a database query that lists the table, the fields to show, the criteria used to filter records and the sort order, written before the query is built.

34Fields, Criteria, Sort

The three parts of a query design: which fields to show, which records match the criteria (using AND/OR), and how results are sorted, so the output meets the requirement.

35spreadsheet mock-up

A sketch of a spreadsheet's worksheet names, header labels, data areas, formatting and result cells, drawn before building to show how it will look and where results will go.

36annotated diagram

A diagram of a database or spreadsheet with labels and notes on its parts, such as tables, fields, data types, cells, formulas and validation, explaining what each part does.

37Interpreting a design

Reading a design's IPO chart, annotations, mock-ups and query designs together to work out what must be built, and checking it matches the requirements.

38primary data vs secondary data

Primary data is data you collect yourself, like a survey. Secondary data was collected by someone else, like data from a website or database.

39authentic data

Real data collected by a trusted organisation and kept in a large repository; check its publisher, date and relevance to your question.

40open data

Datasets, such as government statistics, that anyone can download from a portal. A licence says how you may reuse them, and you must credit the source.

41file format

The way a dataset's data is structured and stored in a file, such as CSV, XLSX, JSON or XML. Choose the format that suits the software you will open it in.

42CSV file

A plain text file that stores table data with commas between the values. A spreadsheet can open it and show the data in rows and columns.

43Import CSV

Fills a new database table from a CSV file. The first row becomes the field names, and you choose a data type for each field, such as text or number, and check them after the import.

44data quality check

Judging a dataset by testing accuracy, bias, integrity, relevance and reliability one at a time, and using evidence to show how any problem could affect conclusions.

45validation

An automatic check that data is reasonable and in the right form when entered. It blocks bad data early, but it cannot prove the data is true (a valid age of 15 could still be wrong).

46validation check

A test that an entry is sensible. Types include presence, data type, range, length, format and list. Each one tests a different property of the entry.

47ISBLANK

A spreadsheet function that checks a cell. It returns TRUE if the cell is empty and FALSE if it holds data, so it works as an existence check. COUNTBLANK counts empty cells in a range.

48ISNUMBER

A function that tests whether a cell value is a number. It returns TRUE for a number and FALSE for text, so it works as a type check. ISTEXT does the same job for text.

49=AND(B2>=0,B2<=100)

A range check formula. It returns TRUE only if the value in B2 is at least the minimum (0) and at most the maximum (100), so boundary values pass and anything outside is rejected.

50Data Validation

A spreadsheet tool that limits cells to allowed entries, like a whole number from 1 to 100 or a drop-down list, and shows an error message if someone types anything else.

51validation rule

A rule set on a database field that checks each entry, such as a number range, a maximum text length or a drop-down list, and rejects any record that breaks the rule.

52CHECK (age BETWEEN 0 AND 120)

A constraint is a rule on a column. This CHECK rejects any record with an age outside 0 to 120. NOT NULL (must have a value) and UNIQUE (no repeats) are constraints too.

53APA referencing

A system that credits the sources of data and information using in-text citations linked to a reference list. It respects intellectual property and helps avoid plagiarism.

54(Smith, 2023)

An APA in-text citation. It puts the author's surname and the year in brackets straight after the information you used, so readers can find the full source in your reference list.

55Author. (Year). Title. Source.

An APA reference list entry gives the author, year, title and source in that order. Entries are listed A-Z by author, with a hanging indent on any lines after the first.

56Publisher. (Year). Title [Data set]. URL

An APA reference for a dataset: publisher, (year), then the title with a [Data set] label, then the URL. For census data the publisher is the Australian Bureau of Statistics.

57Author. (Year, Month Day). Title. Site Name. URL

An APA web page reference lists the author or organisation, date, page title in italics, site name and URL. If no date, use (n.d.); with no author, start with the title. URL goes last.

58OpenAI. (2024). ChatGPT (GPT-4o) [Large language model]. https://chatgpt.com

An APA reference for AI-generated content: the tool's developer as author, then year, tool name and version, a [Large language model] label and the URL.

61fill handle

The small square at a cell's corner. Drag it down to copy a formula to the cells below. The cell references change row by row, so B2 becomes B3.

63=SUM(B2:B10)

A formula that adds up all the numbers in a range of cells. The colon means 'from B2 to B10'. The AutoSum button can type it for you.

64=AVERAGE(B2:B10)

A spreadsheet function that adds up the numbers in cells B2 to B10 and divides by how many there are, to find the mean (the typical value). It is not the total.

65=MIN(range) and =MAX(range)

MIN finds the smallest (lowest) value in a range of cells and MAX finds the largest (highest) value. For example, =MAX(B2:B10) shows the top score in cells B2 to B10.

66=COUNT(B2:B10)

A spreadsheet formula that counts how many cells in the range B2 to B10 contain numbers. It ignores cells with text or empty cells.

67=COUNTIF(range, criteria)

Counts how many cells in a range match a value or condition, such as "Yes" or ">10". Example: =COUNTIF(C2:C30,"Yes") counts every Yes in C2 to C30.

68=SUMIF(range, criteria, sum_range)

Adds only the numbers in sum_range whose matching cell in range meets the criteria, such as the total of all sales for one category.

69=IF(test, value_if_true, value_if_false)

Checks a test, like B2>=50. Shows the second value if the test is true and the third if it is false. Text results go in quotes, e.g. =IF(B2>=50,"Pass","Fail").

70XLOOKUP

A lookup function that finds a key in one column and returns the matching value from another column, such as a price for a product code. It can show a message if no match is found.

71Sort

Sorting puts the rows of a table in order using the values in one column, like A to Z or smallest to largest, and keeps each row together.

72filter

A filter shows only the rows that match a value you choose and hides the other rows. The hidden rows are not deleted. Clear the filter to see them all again.

73clean data

Finding and fixing mistakes in data, like typos, spelling variations, duplicate rows, missing values and impossible values, so the results are correct.

74=TRIM(A2)

A text function that returns a tidy copy of the text in A2 with extra spaces removed. UPPER, LOWER and PROPER work the same way to fix capitals, e.g. =PROPER(A2) gives Mia Chen.

75Remove Duplicates

A spreadsheet tool that deletes rows repeating the same values in the columns you choose to compare, keeping the first match. Always keep a copy of the raw data first.

76missing value

An empty cell where data should be. Find it with a filter for blanks, then delete the row, fill it from a reliable source or exclude it so AVERAGE and COUNT are not distorted.

77=VALUE(A2)

Converts text that looks like a number, such as '45' stored as text, into a true number so it can be added, averaged and sorted correctly. Left-aligned numbers are a warning sign.

78=LEFT(A2,3)

Text functions split text or join it. LEFT, RIGHT and MID take part of a value; CONCAT or & joins parts into one. This formula returns the first 3 characters of cell A2.

79cell formatting

Changing how a cell looks, such as currency, decimal places, bold or column width, without changing the value stored inside it. For example, 5 can show as $5.00.

80error value

A message starting with # (such as #DIV/0!, #N/A, #VALUE! or #REF!) that a cell shows when a formula fails. It hints at what went wrong so you can fix the formula or data.

81=MEDIAN(B2:B30)

Returns the middle value of the numbers in B2:B30 once ordered. With an even count it averages the middle two. You do not need to sort the cells first.

82mean vs median

An outlier drags the mean towards it but barely moves the median. If the two differ a lot, the data is skewed, so the median is usually the fairer typical value to report.

83=MAX(B2:B30)-MIN(B2:B30)

The range is the largest value minus the smallest value. It measures spread, and one outlier can make it much bigger.

84=COUNTIF(range,A2)

Used in a frequency table: list each category in column A, then put this beside it to count how many times that category appears in the data range. Copy it down for every category.

85=COUNTIFS(B2:B30,">=10",B2:B30,"<20")

Counts the numbers in a range that are at least the lower limit and less than the upper limit, giving the frequency of one bin. Equal-width bins must not leave gaps or overlap.

86=STDEV.S(B2:B30)

Returns the sample standard deviation of the numbers in a range, showing how spread out they are. Use STDEV.P if the range is the whole population.

87standard deviation

A measure of how spread out values are around the mean. A small value means the data is clustered close to the mean. A large value means it is widely spread. It uses the same units as the data.

88=CORREL(B2:B9,C2:C9)

Finds Pearson's r for two equal-length number ranges. The result is from -1 to 1: near 1 is strong positive, near -1 is strong negative, near 0 is no linear link.

90scatter graph

A chart that plots each record as a point using two number columns, one on the x axis and one on the y axis, to show whether two things are related.

91trendline

A line fitted through a chart's points that shows the overall direction of the data. A chart tool can add one with its equation, and the R-squared value shows how well it fits.

92correlation is not causation

Two things can rise or fall together without one causing the other. A trend shows a relationship, not the reason, so a hidden third factor may explain both.

93histogram

A chart with touching columns showing how many values fall in each numeric bin. Bins go on the x axis, frequency on the y axis. It shows the shape of the data, unlike a bar chart of categories.

94distribution shape

How data values are spread out, seen on a histogram: symmetrical (bell-shaped or uniform) or bimodal, which has two peaks. Skew is covered separately.

95skew

Lopsidedness of a distribution: positive skew has a long tail to the high side, negative skew to the low side. The mean is pulled towards the tail, away from the median.

96pattern, trend, outlier

Look for patterns and trends in data, like highest, lowest, going up or down. An outlier is a value that stands out from the rest. Compare groups to see differences.

97conclusion

A statement that answers a question using evidence from the data, such as numbers. It should only say what the data shows and stay within the data's limits.

98exploratory data analysis

Using quick charts and summaries to discover patterns, relationships and outliers in data. The analyst is the audience, trying different charts to form questions.

99presentation visualisation

A visualisation that clearly shows findings to an audience, with a polished design that makes the key message obvious.

100interactive visualisation

A visualisation where users explore the data themselves, for example by filtering, selecting items or hovering over points for more detail.

101purpose and audience

The reason a visualisation is made and who will view it. Together they decide the content, level of detail, tone and style, so the same data can look different for different audiences.

102long-form infographic

A tall, single-page visualisation that tells one story from start to finish as you scroll down through headed sections of text, graphics and charts.

103poster series

A set of separate infographic posters that share one template and colours, each with one message, linked by a common theme.

104dashboard

A single screen that combines charts, tables and key figures so a viewer can monitor data at a glance, such as sales or attendance.

105static vs interactive vs live

A static visualisation is fixed. An interactive one lets users change the view. Live data updates automatically, while stored data is a fixed copy saved at one point in time.

106dynamic data visualisation

An interactive visual linked to live data, so charts update when the data changes or the user refreshes it or uses controls like filters.

107pivot table

A tool that summarises a big table by grouping rows under row and column fields, then working out a sum, count or average for each group. Refresh it when the data changes.

108slicer

A set of buttons linked to a table, pivot table or chart. Click a button to filter the data, so a viewer can explore it interactively. Clear the filter to show everything again.

109Mock-up

A static visual representation of a solution's interface, showing layout and design before development begins

110Data visualisation mock-up

A sketch of a chart or graph made before building it, showing the chart type, titles, labels, colours, fonts and layout so the design can be checked and improved.

111Storyboard

A sequence of visual panels showing how a user moves through and interacts with a solution, step by step

112Choosing the right chart

Pick the chart that matches your question: a bar chart to compare categories, a line chart for change over time, and a pie chart for parts of a whole.

113bar chart

A chart that uses bars to compare amounts in different categories. Select the labels and values, then choose Insert chart. Column charts have upright bars; bar charts lie sideways.

114line chart

A chart that joins points with a line to show how a value changes over time. Time goes along the bottom axis, so you can see if the trend goes up or down.

115pie chart

A circle chart split into slices. Each slice shows how big one part is compared with the whole, often as a percentage. All the slices add up to 100%.

116Title, axis labels and legend

The parts that help a viewer read a chart: the title says the topic, axis labels name the data and units, the legend explains colours.

117Text and graphics

Titles, headings, captions and key figures (text) plus shapes, pictures and icons (graphics) that frame and explain the data so viewers know what it shows and what matters.

118table

A visualisation component that shows exact values in rows and columns, used when the audience needs the precise numbers. Clear headers and only the rows that matter keep it readable.

119chart

A visualisation component that turns data into shapes such as bars, lines or slices so patterns can be seen quickly. Use one chart per message and place it near its explanation.

120colour palette

A small, consistent set of colours where each colour has one meaning. Use the same colour for the same category, and never rely on colour alone, so colour-blind viewers can still read it.

121font hierarchy

Using only two or three fonts, with size and weight showing importance from title to heading to body to caption, so a visualisation is easy to read.

122icon

A small simple picture that makes a topic quick to recognise. Keep icons relevant and in one style, use them only where they help the data, and credit or get permission for any you borrow.

123consistent design

Using the same colours, fonts, alignment and spacing on every part of a dashboard or poster series so it looks unified and is easy to read.

124visual hierarchy

Using size, position, colour and weight to show which element viewers should read first, so the key message stands out and the reading order is clear.

125clarity of message

A visualisation has clarity of message when a viewer can state its one key point quickly, because clutter is removed and a descriptive title tells them what to look for.

126Conditional formatting

A spreadsheet tool that changes a cell's colour or icon automatically when its value meets a rule, so highs, lows and outliers stand out.

127misleading chart

A chart whose design, like a cut-off axis, 3D effects or a cherry-picked range, makes the data look different from what it truly shows.

128Justify a design choice

Give a reason for a design choice (chart, colour, format) that links it to the purpose of the solution, the needs of the audience and the type of data. Use 'because' to link them.

129Machine learning

AI learns patterns from large amounts of training data, then uses those patterns to make predictions or generate content.

130built-in AI features

Smart tools inside everyday software, like Copilot in Excel, that suggest formulas and charts, answer plain-English questions and find insights automatically.

131visualisation prompt

An instruction to an AI tool that names the data fields, chart type, title, axis labels, colours and audience so it can build a clear, suitable chart.

132Refining a prompt

Reviewing the AI's chart, then sending a follow-up prompt that names exactly what to change, such as labels, chart type or detail, and repeating until it is right.

133Check against source data

Compare every value, label and axis scale in an AI-generated chart with the original data, and look for invented figures, before you trust or use the chart.

134AI-generated data

New data, such as text, images or predictions, created by an AI system. Its quality depends on the data the AI was trained on, so students must verify it before use.

136statistical model

A rule built from past data, like a regression line, that uses the relationship between variables to estimate new values. A strong r makes the estimates more reliable.

137=FORECAST(x, known_ys, known_xs)

Estimates a new y value for a given x by extending the trend of existing data, e.g. predicting next year's sales. It is less reliable the further x is beyond the known data.

138model

A model uses data to predict what will happen, support decisions and recommend options to users, such as suggested products.

139reliability

How much you can trust a prediction. It drops with small samples, outliers and predicting far beyond your data, so always state how uncertain the prediction is.

140biased data

Data that unfairly favours one view or group because of how it was collected or selected, e.g. sampling bias, leading questions or missing groups.

141normal, boundary and invalid data

A good test set uses normal data (typical values), boundary data (values at the edges of a range) and invalid data (values that should be rejected), so every path in an algorithm is checked.

142Boundary values

Test inputs at the edge of a condition. For age >= 18, test 17 (just below), 18 (exactly on) and 19 (just above) to check the algorithm picks the right path at the edge.

143test case

A chosen input paired with the output you expect. Run the algorithm, record the actual output, then compare: if actual matches expected it passes, otherwise it fails.

144expected result

The answer you work out by hand using a few rows of data. Compare it with the formula's actual result. If they match, the formula works. If not, check the formula and its range.

145Test a query

Predict the rows a query should return from a small sample, run it, then compare the actual result with your prediction. Check row counts, boundary values and JOIN matches.

146test data

Values chosen to test each validation rule: empty for existence, text for type, and values just inside and just outside range limits, each with an expected result.

147visual inspection

Checking the appearance of a visualisation, such as layout, alignment, readability, colours and labels, often with a checklist to spot problems.

148Chart verification

Comparing a chart's plotted values, category labels, totals and axis scale with the source data to confirm the chart shows the data accurately.