Memra
academic · beginner

AFM 112 — Analytic Methods for Business 1

The data-analytics process from Sheets to Excel to R, on one company’s data — 90 lessons

The whole of AFM 112 as its open-book quizzes and final test it: how the course is assessed, the CRISP-DM process and how it serves the SAF problem-solving process, data and variable types, spreadsheet formula mechanics, descriptive statistics, data preparation with logical and lookup formulas, pivot tables, charts and visualisation, then the R half — the on-ramp with tiny vectors, the same twelve-row orders data frame the learner already knows from Sheets, data understanding and preparation in base R, simple models and how to read summary(lm()), and dashboards, deployment and the crew project. One company, Maple & Birch Office Supply, runs through all fourteen modules so the same numbers recur in both tools; every scenario question is multiple choice with the adjacent wrong tool as the distractor; every line of R output came from a real run.

Start course → Flashcards 755
0 / 90 lessons
We'll stop scheduling reviews for after it.

How This Course Is Examined, and the CRISP-DM Process

49 cards
  1. Open-book, online, and the final is the shock 7 min ◈ 5
  2. What business analytics is, and who it serves 7 min ◈ 5
  3. Six phases, arrows both ways, an outer loop 8 min ◈ 6
  4. Business understanding: objectives, situation, data goal, plan 9 min ◈ 6
  5. The three phases where the tools live 9 min ◈ 7
  6. Business fit vs technical fit, and what "deploy" means here 8 min ◈ 7
  7. Client-facing stages, data-facing phases 8 min ◈ 5
  8. The primary MCQ skill, drilled 9 min ◈ 8

Data: Records, Variables, and Types

40 cards
  1. Rectangular data, sources, and the tidy layout 8 min ◈ 5
  2. The first split, and the numeric-looking trap 8 min ◈ 6
  3. Nominal, ordinal, interval, ratio — as permitted operations 9 min ◈ 7
  4. The describe-data deliverable 8 min ◈ 5
  5. Three types that misbehave 10 min ◈ 9
  6. Blank, NA, −999, and why they must be declared 8 min ◈ 8

The Spreadsheet Grid and Formula Mechanics

33 cards
  1. Addresses, ranges, and the fill handle 7 min ◈ 8
  2. `=`, operators, parentheses — and the same rule in R 8 min ◈ 8
  3. `$` and what the fill handle does to each 10 min ◈ 9
  4. A stable source, and checking your own formulas 8 min ◈ 8

Describing Data in a Spreadsheet

68 cards
  1. Counting a categorical variable, three ways 9 min ◈ 10
  2. The centre, in both tools 9 min ◈ 12
  3. Which centre to report for money-like data 8 min ◈ 6
  4. The middle half and the 1.5·IQR rule 10 min ◈ 10
  5. Spread, sample vs population, and comparing across scales 10 min ◈ 11
  6. From a numeric column to a distribution 10 min ◈ 9
  7. Which cell answers which question 9 min ◈ 10

Data Preparation I: Derived Variables, Text, Dates, and IF

55 cards
  1. One formula per row in Sheets, one line in R 9 min ◈ 11
  2. TRIM, UPPER, LEFT, & — and their R twins 9 min ◈ 10
  3. Days between, month buckets, and the epoch trap 8 min ◈ 7
  4. IF and ifelse, and the case-sensitivity conflict 9 min ◈ 8
  5. Compound conditions: functions in Sheets, operators in R 8 min ◈ 7
  6. Three-plus bands: IFS, nested IF, and cut() 10 min ◈ 12

Data Preparation II: Conditional Aggregates, Lookups, Sort and Filter

73 cards
  1. Criteria are text, wildcards work, cells need & 9 min ◈ 9
  2. The sum range moves first in the plural form 10 min ◈ 11
  3. Lookup value, table, column index, match type 10 min ◈ 9
  4. FALSE for IDs, TRUE only for a sorted band table 9 min ◈ 8
  5. The modern lookups, and what each error means 9 min ◈ 9
  6. Keys, inner vs left, unmatched rows 9 min ◈ 9
  7. Records stay intact; filters hide, they do not delete 9 min ◈ 9
  8. Every M5–M6 operation, tagged 9 min ◈ 9

Pivot Tables: Aggregating and Comparing Groups

55 cards
  1. The four drop zones 8 min ◈ 6
  2. One category → Rows; two → Rows × Columns; "for only…" → Filter 9 min ◈ 7
  3. Why is my total 12? 8 min ◈ 7
  4. "Of Ontario's sales" vs "of online sales" 10 min ◈ 9
  5. Months from dates, and the two ways to filter 8 min ◈ 8
  6. A cell is the summary of the rows matching both labels 9 min ◈ 7
  7. Pivot ⇔ SUMIFS ⇔ aggregate / tapply 10 min ◈ 11

Charts and Visualisation Principles

43 cards
  1. A chart with no message is a table in disguise 7 min ◈ 5
  2. Bar for categories, histogram for a numeric, line for time, scatter for two numerics 9 min ◈ 8
  3. The time axis is categorical; stacked hides everything but the bottom 9 min ◈ 6
  4. The "which display is inappropriate" item 7 min ◈ 6
  5. Zero-based bars, honest time axes, index to 100 9 min ◈ 9
  6. Comparison, distribution, trend, relationship, composition 9 min ◈ 9

R and RStudio I: The Console, Vectors, and Types

56 cards
  1. Reproducible, re-runnable, auditable 8 min ◈ 8
  2. [1] is an index, not a value 8 min ◈ 8
  3. A vector is a column 9 min ◈ 10
  4. One text entry turns the whole column chr 8 min ◈ 8
  5. Position, range, drop, condition — and R is 1-based 9 min ◈ 10
  6. ?mean, defaults, and the first R gotcha 10 min ◈ 12

R and RStudio II: Factors, Data Frames, and Reading Data

50 cards
  1. Levels, and why their order matters 8 min ◈ 8
  2. Equal-length columns, $, nrow, dim, names 9 min ◈ 10
  3. "cannot open file" means the wrong folder 9 min ◈ 8
  4. str() is the data dictionary check 9 min ◈ 8
  5. Four messages, four causes 8 min ◈ 7
  6. Index prefixes, row names, truncated frames 8 min ◈ 9

Data Understanding in R

56 cards
  1. Min · 1st Qu. · Median · Mean · 3rd Qu. · Max · NA's 9 min ◈ 9
  2. Counts, shares, margins 9 min ◈ 9
  3. df[rows, cols], subset(), and the trailing comma 10 min ◈ 11
  4. &, |, !, %in%, and the proportion idiom 9 min ◈ 9
  5. order() keeps records together; which.max is a position 8 min ◈ 8
  6. Look, then check 9 min ◈ 10

Data Preparation in R

67 cards
  1. df$new <- …, names(), <- NULL 8 min ◈ 9
  2. Two-way flags and bands, vectorised 9 min ◈ 8
  3. as.numeric, as.Date(format =), gsub 9 min ◈ 9
  4. na.omit, imputation, duplicated 9 min ◈ 10
  5. The pivot, as a formula 10 min ◈ 11
  6. Join, stack, and the named-vector lookup 9 min ◈ 10
  7. Load → inspect → clean → derive → aggregate → write 10 min ◈ 10

Simple Models: Scatter, Correlation, the Line, and Reading lm()

64 cards
  1. Response on the vertical axis, chosen from the decision 8 min ◈ 7
  2. Sign, strength, linear only, and never causation 9 min ◈ 10
  3. ŷ = b₀ + b₁x, residuals, least squares 10 min ◈ 11
  4. RSQ, FORECAST.LINEAR, the trendline, the ToolPak block 9 min ◈ 8
  5. Estimate, Multiple R-squared, Residual standard error 11 min ◈ 10
  6. Plug in, but only inside the x-range 9 min ◈ 9
  7. Sign, size, range, decision — and a pivot is a model too 9 min ◈ 9

Communicating Results: Dashboards, Deployment, and the End-to-End Walk

46 cards
  1. One message per panel; six panels at most 8 min ◈ 6
  2. Data sheet → pivots → pivot charts → slicer; R feeds the data sheet 9 min ◈ 9
  3. Same colour per measure, zero-based bars, per-unit not totals 8 min ◈ 7
  4. Question stated? Right chart? Right aggregation? Fair comparison? Filters obvious? Units and dates? 9 min ◈ 7
  5. Who refreshes it, did it meet the criteria, how a crew ships 9 min ◈ 8
  6. The final's signature item 11 min ◈ 9
NORMAL ~/memra/learn/afm-112 utf-8 LF