# How to analyze assessment data faster on Mac: a teacher's routine with Numbers 15.4

> Turn quiz and assessment results into next steps in less time. A Mac routine for Numbers, Google Forms and Microsoft Forms, plus honest Lazur notes.

URL: https://www.lazur.app/for/teachers/problems/analyze-assessment-data-faster-teachers
Updated: 2026-11-09

You analyze assessment data faster when you stop reading scores one student at a time. Export the results to one sheet, sort students into groups, then write one next step per group. This page walks through that routine on a Mac, with export steps for Google Forms and Microsoft Forms and an honest look at where Lazur helps.

## Quick answer

- Export results to one spreadsheet before you look at a single score.
- Sort students into groups by next step, not by total score alone.
- Pick one next step per group and put a date on it.
- Google Forms gives you a CSV file. Microsoft Forms opens results in Excel.
- Lazur explains what is on screen and drafts text. It does not export or click for you.

## What turning assessment results into next steps costs you

- **Hours lost to cleaning exports.** Before you can read a single score, you may need to fix the file. Names differ between the roster and the form. Columns arrive in an order you did not expect. Each fix looks small, but they stack up across every quiz in a term. Budget the cleaning time as part of the job.
- **Feedback that arrives after the lesson moved on.** A group that needed reteaching on Tuesday gets it the next Monday, if at all. By then the class has covered new material, and the gap has grown. Late analysis costs the students who needed help most. The fix is a fixed slot in the same week as the quiz.
- **Reteaching the whole class by default.** Without groups, the easy answer is to reteach everything. That spends class time on skills most students already have. Students who got it wait while you cover the basics again. Sorting students into groups costs a few minutes. Reteaching everyone costs a whole lesson.
- **Evenings and weekends.** The BLS puts the May 2025 median pay for kindergarten and elementary school teachers at $63,820, so unpaid hours are a real cost. Analysis you do at home is time you do not get back. Where you can, protect school hours for it.
- **Decisions made on the wrong evidence.** A total score can send you to the wrong fix. If the class missed one question, a full reteach wastes time. If a few students missed many questions, you need a different response. Without the item view, you guess, and guessing costs the time you were trying to save.

## Why it happens

1. **Results sit in different places.** Quiz results come from paper, a learning platform, Google Forms, Microsoft Forms and email. Each source exports differently. On a Mac you may use Numbers, Google Sheets and Excel in the same week. Without one place to combine them, the analysis never starts.
2. **Totals hide the skill gap.** A total score shows who is low. It does not show why. Microsoft Forms exports points and feedback for each question. Google's export depends on how the quiz is set up. Use the question columns when they exist. Otherwise, the next step is a guess.
3. **Nobody protects the time.** EdWeek's breakdown lists grading and feedback, lesson planning and general administrative work. Analysis is not a separate line. When a task has no slot, it loses to the tasks that do. Analysis is easy to skip, because nothing visibly breaks when you skip it.
4. **Acting on data is a skill that needs practice.** Shanahan and colleagues found that ongoing support strongly improved teachers' data-based skills, with g = 0.86 on teacher outcomes. Reading data and acting on it takes repeated practice. The review studied ongoing support, not one-off sessions.
5. **Assessment routines that feel pointless.** A 2001 survey of 1,001 teachers in England and Wales found nine out of ten spent a lot of time on assessment procedures most considered ineffective. The survey was commissioned by an assessment provider, and it is old. Still, the feeling is familiar. When the output goes nowhere, the work feels like paperwork.

## How to fix it

- **Export once into one sheet.** Pick the file type your form gives you. For Google Forms, use Responses tab > More > Download responses (.csv). For Microsoft Forms, use Responses tab > Open in Excel. Open the file in Numbers, and check your version under Numbers > About Numbers. Keep one file per quiz, named with the unit and the date.
- **Sort by the question, not only the total.** Find the question that most students missed. Then list the students who missed it. Two short lists tell you more than a sorted total. Use the question column and the points column. Do not rely on memory for which question was hard.
- **Make three groups, no more.** Use three labels: secure, nearly there, not yet. Three groups are easy to remember and easy to teach from. More groups usually mean one-student groups you cannot run. Put the label in a column called Next step, so it is easy to find later.
- **Write one next step per group, with a date.** A good next step names one skill, one group and one check. For example: reteach common denominators with two worked examples, then check with three problems on Friday. The date turns a plan into a task. Without it, the step drifts until the unit ends.
- **Let students act on results too.** Xuan and colleagues found formative assessment worked better when teachers and students both acted on results. Give students their own result sheet with one target. Ask them to mark one next step. The pooled effect in that review was about 0.19 on reading, so treat it as one tool among several.
- **Keep one template and never rename columns.** Use the same column names every term. Copy the sheet, clear the scores, and keep the headers. Last year's data then lines up with this year's, so you can compare groups over time. Name the reused file by unit, not by date.

## Why turning results into next steps takes so long

Collecting results is quick. Acting on them is slow. The slow part sits between the export and the lesson you teach next. You read a score, guess the reason, and move on. Then you repeat that for every student and every quiz.

Most of that time hides inside tasks nobody names. The 2022 EdWeek breakdown put the median teacher at 54 hours a week. Grading and feedback took 5 of those hours. The article does not state the survey year or sample size, so read these as one snapshot, not a fixed rule.

- **54** Median hours per week US teachers work, per a 2022 EdWeek survey ([source](https://www.edweek.org/teaching-learning/how-teachers-spend-their-time-a-breakdown/2022/04))
- **5** Hours per week on grading and feedback in the same EdWeek breakdown ([source](https://www.edweek.org/teaching-learning/how-teachers-spend-their-time-a-breakdown/2022/04))
- **0.86** Effect size on teacher data-use outcomes from ongoing support (Shanahan et al., 2024) ([source](https://pmc.ncbi.nlm.nih.gov/articles/PMC11636021/))
- **0.31** Effect size on student outcomes from the same ongoing support meta-analysis ([source](https://pmc.ncbi.nlm.nih.gov/articles/PMC11636021/))

The same breakdown has no line for analyzing assessment data. Analysis hides inside grading, planning and general admin. That is why it rarely gets a slot of its own. If you do not name it, you will not protect it.

## The step most teachers skip: deciding what the results mean

Grading tells you what each student got right. Analysis tells you which skill is missing, and for whom. Those are different jobs. A total score cannot tell you whether the miss came from the math or from reading the question.

Look for patterns across the class. Ask which question most students missed. Then ask which students missed that same question. Both checks take minutes once the data sits in one sheet, and both point at a next step.

![Teacher at a laptop in a classroom, using a spreadsheet to analyze assessment data faster](https://images.lazur.app/content-images/problem/analyze-assessment-data-faster-teachers/analyze-assessment-data-faster-1-95a247.jpg)

## A worked example with a fractions quiz

Say your class took a short fractions quiz on Friday. On Monday you export the results. Each row is one student. Each column is one question, plus a total. Your first job is to find the question that caused the most trouble.

Suppose question four asks students to add two fractions with different denominators. Most of the class misses it. Only a few students miss question one. The next step is clear. Reteach common denominators to the whole group, not the whole quiz.

Now take the students who missed question four. Split them into two groups. One group needs a worked example. The other needs a quick check on the same skill before they practice. Each group gets one next step and a day to check it.

## The routine, in order

1. Export the responses as a CSV or Excel file on Monday.
2. Open the file in Numbers, or in Google Sheets in your browser.
3. Add a column called Next step, and leave it blank for now.
4. Sort students into three groups: secure, nearly there, not yet.
5. Write one next step per group, with the day you will check it.

## Export quiz results on Mac from Google Forms and Microsoft Forms

Google Forms exports from the Responses tab. Click More, then Download responses (.csv). Google's help page also describes Link to Sheets, which keeps a live spreadsheet connected to the form. The page does not say which quiz fields the file includes. Open the file and check the columns yourself.

Microsoft Forms works differently. On the Responses tab, select Open in Excel. The workbook has one row per response and one column per question. It includes start and completion times, total points, and points and feedback for each question. If the quiz is anonymous, the name and email columns are left out.

| Form tool | Export path on Mac | What the file holds |
| --- | --- | --- |
| Google Forms | Responses tab > More > Download responses (.csv) | One row per response. Google's help page does not confirm quiz fields, so check the columns. |
| Microsoft Forms | Responses tab > Open in Excel | One row per response and one column per question, with points and feedback for each question. |

_Two ways to get quiz results into a spreadsheet on Mac._

Keep the export as a plain copy. Do your analysis in that copy, not in the form's own view. Save the file with the unit name first, so you never mix two quizzes in one sheet.

## Which Mac app to use for the analysis

Numbers suits this job well. Apple's Numbers guide covers Numbers 15.4. To check your own version, choose Numbers > About Numbers from the menu bar. Google Sheets in your browser works too, and so does Excel if your school provides it. The app matters less than the habit.

Pick one tool and use it for every quiz. Switching apps each term means relearning the sort and the filter. It also means old files stop matching new ones, so comparisons break.

## Turning groups into next steps you will actually do

A next step should name one skill, one group and one check. Reteach common denominators with two worked examples is a next step. Work on fractions is not. Write the action, the group, and the day you will check it.

Keep the note where you will see it on Monday. A planning document in [Google Docs](https://www.lazur.app/for/teachers/google-docs) works well for this. A page in Pages works too, if you prefer its layout.

> **Keep each step small:** If a next step takes more than one lesson, split it. One skill, one group, one check. Small steps are easier to finish and easier to judge a week later.

## What the research says about acting on results

The research is clearer on teacher skills than on student results. In a 2024 meta-analysis, Shanahan and colleagues studied ongoing support for data-based individualization. The effect on teacher outcomes was g = 0.86. The effect on student outcomes was g = 0.31.

The g value is a standardized effect size. Bigger means a larger effect. The student figure is the modest one, so support helped teachers read data far more than it moved scores directly. Student gains still counted. The authors' discussion reports slightly different numbers, so check the paper before you quote it.

Xuan, Cheung and Sun (2022) reviewed formative assessment in K-12 reading. Across 48 studies and 116,051 students, the pooled effect was about 0.19. The stronger finding came from integration. Assessment that teachers and students both act on beat teacher-directed assessment alone.

## Privacy and school rules come before any tool

> **Check student records rules first:** Your school may restrict AI tools on screens that show names, IDs or grades. Lazur sends the audio and the screenshot to AI providers to answer. Lazur says they are not stored.

In practice, hide names with initials or student IDs. Test on a dummy file first, with made-up names. Keep Lazur on your own question, not on a roster. Lazur listens and looks at the screen only while you hold its shortcut, ⌥ ⌘.

## Mistakes that slow the analysis down

- Scoring again before you have a plan for the scores.
- Keeping two exports and comparing them by eye.
- Renaming columns each term, so old and new files stop matching.
- Grouping students by total score only, which hides the skill gap.
- Writing next steps that nobody reads on Monday.
- Skipping the check, so you never learn whether the step worked.

The last mistake matters most. A next step without a check is a guess. Put one short check on the calendar for the lesson after the reteach. Use a few questions or an exit ticket. Then compare the group's results with the first quiz.

## Where Lazur fits, and where it does not

Lazur is a voice assistant for your Mac. Hold ⌥ ⌘ to ask about anything on screen. It answers out loud and points at things. Ask where Download responses sits, and Lazur circles it. Ask how to link a sheet, and it guides you one click at a time.

It does not export your quiz, sort your column or click for you. It cannot make the call about a student. Treat its reading of a table as a second look, then check the rows yourself. That is the honest split of work.

Hold fn to dictate your next steps at the cursor in any app. A personal dictionary helps with test names and unit titles. Dictation is included, so notes can go straight into your planning document.

## A realistic week with this routine

The plan below fits one unit test. You do not need a new meeting or a new app. You need one fixed block and a short list of groups. Run it once with a real quiz, and time each step. Those times tell you what to change next.

![Teacher marking student work at a desk, the grading and feedback time that eats the week](https://images.lazur.app/content-images/problem/analyze-assessment-data-faster-teachers/analyze-assessment-data-faster-2-01e178.jpg)

## Where the evidence is thin

Be clear about the gaps. No open study in this review measures how many teachers struggle to turn results into next steps. The EdWeek breakdown does not split out analysis time. A 2001 England and Wales survey found nine out of ten teachers spend a lot of time on assessment procedures they consider ineffective.

That survey was sponsored by an assessment provider and is now old. Use it as a hint, not a benchmark. So the routine above rests on sound logic and modest evidence. It costs almost nothing to try for one unit.

## What to do next

Pick one class and one upcoming quiz. Run the routine once and time it. Save the sheet as your template, and keep the column names. Next term, copy the sheet and change only the questions.

For feedback letters, see [how to add a comment in Pages](https://www.lazur.app/how-to/add-a-comment-in-pages). To chart how many students sit in each group, see [how to add a chart in Pages](https://www.lazur.app/how-to/add-a-chart-in-pages). When you are ready to try Lazur, the [download page](https://www.lazur.app/download) has the app.

![Person working at a laptop on a desk, a setting for a short weekly analysis block](https://images.lazur.app/content-images/problem/analyze-assessment-data-faster-teachers/analyze-assessment-data-faster-3-ef30aa.jpg)

## With Lazur

Lazur is a voice assistant for the Mac that lives in the notch. Hold ⌥ ⌘ and ask; it sees your screen, answers out loud, points at buttons and walks you through apps. Hold fn to dictate. Free for 3 days, then $18/month: https://www.lazur.app/download

- “Where is the CSV download on this page?”: Lazur points at the option and labels it. You make the click yourself. (Points at it)
- “Hold ⌥ ⌘ and ask which question most students missed.”: Lazur reads the table on screen and answers out loud. Check its answer against the rows before you use it.
- “Where is the Download responses option?”: Lazur circles the option and labels it on screen. You make the click yourself.
- “Walk me through Link to Sheets, one click at a time.”: Lazur guides you one click at a time and waits for each click before it moves on.
- “Hold fn and dictate the next step for each group.”: Dictation types clean text at the cursor in any app, so the note lands in your planning document.
- “Draft a short note to the nearly there group about Friday's check.”: Lazur drafts the text. You read, edit and send it. Lazur does not send email for you.
- “Remind me on Monday to export the quiz.”: Reminders and Calendar events work by voice, so the export step gets a time on your calendar.

## A plan you can start this week

- **Monday: Export and save the file.** Close the quiz. Export the responses as a CSV, or open Microsoft Forms results in Excel. Save the file with the unit name and the date. Do not edit it yet.
- **Tuesday: Clean the names.** Check that each name matches the roster. Remove test responses and blank rows. Keep the column headers exactly as the form gives them, so next term's file lines up.
- **Wednesday: Find the missed questions.** Scan the question columns. Write down the one question most students missed. Then list the students who missed it. Keep that list in a separate note for your planning document.
- **Thursday: Make the groups and next steps.** Add the Next step column. Label each student secure, nearly there or not yet. Write one next step per group, with the day you will check it.
- **Friday: Teach and check.** Teach the reteach lesson to the group that needs it. Run a short check the same day or the next Monday. Note who moved up and who did not.
- **Next Monday: Review the change.** Compare the check with the first quiz. Decide which next step worked, which needs another try, and which can stop. Save the sheet as your template for the next unit.

## FAQ

### How do I analyze assessment data faster as a teacher?

Export everything to one spreadsheet, then look at the question columns before the totals. Sort students into three groups by next step. Write one action per group with a date. Doing this once for a real quiz shows you where your time goes.

### Can Lazur analyze my assessment spreadsheet for me?

Lazur can read what is on your screen when you hold ⌥ ⌘, and it answers out loud. It can point out patterns it sees. It does not sort, export or edit your file, and it does not click for you. Check its reading against the rows.

### Is it safe to use Lazur with student names on screen?

Check your school's rules first. Lazur sends the audio and the screenshot to AI providers to answer. Lazur says they are not stored. Hide names with initials or IDs, or use a dummy file, if your school limits AI tools.

### Where do I find quiz results in Microsoft Forms on Mac?

Open the form, then select the Responses tab. Choose Open in Excel to get one row per response and one column per question. Use Review Answers to grade. Use Post scores to release scores and feedback to students.

### Does the Google Forms CSV include quiz feedback?

Google's help page for responses does not describe quiz-specific export details. Download the CSV, open it in Numbers, and check whether points and feedback columns are there. Do not assume they are.

### Is there good evidence that this routine works?

The evidence is thin on this exact task. A 2024 meta-analysis found a large effect on teacher data skills and a smaller effect on student outcomes. Try the routine on one unit, then judge it by time saved and by whether your checks improve.

### What does g = 0.86 mean?

It is a standardized effect size from a meta-analysis, where bigger means a larger effect. In the 2024 review, ongoing support had g = 0.86 on teacher outcomes and g = 0.31 on student outcomes. Check the paper, because its discussion reports slightly different figures.

## Sources

1. [How Teachers Spend Their Time: A Breakdown](https://www.edweek.org/teaching-learning/how-teachers-spend-their-time-a-breakdown/2022/04), Education Week, checked 2026-11-09
2. [Time Spent On Assessment 'Ineffective'](https://www.ipsos.com/en-uk/time-spent-assessment-ineffective), Ipsos, checked 2026-11-09
3. [Ongoing Teacher Support for Data-Based Individualization: A Meta-Analysis and Synthesis](https://pmc.ncbi.nlm.nih.gov/articles/PMC11636021/), PubMed Central (NIH), checked 2026-11-09
4. [The effectiveness of formative assessment for enhancing reading achievement in K-12 classrooms: A meta-analysis](https://pmc.ncbi.nlm.nih.gov/articles/PMC9443994/), PubMed Central (NIH), Frontiers in Psychology, checked 2026-11-09
5. [Kindergarten and Elementary School Teachers, Occupational Outlook Handbook](https://www.bls.gov/ooh/education-training-and-library/kindergarten-and-elementary-school-teachers.htm), U.S. Bureau of Labor Statistics, checked 2026-11-09
6. [Numbers User Guide for Mac](https://support.apple.com/guide/numbers/welcome/mac), Apple Support, checked 2026-11-09
7. [Check and share your quiz results](https://support.microsoft.com/en-us/forms/check-and-share-your-quiz-results), Microsoft Support, checked 2026-11-09
8. [View & manage form responses](https://support.google.com/docs/answer/139706?hl=en), Google Docs Editors Help, checked 2026-11-09