The advice arrived in a first meeting, kindly given by someone with more years in the field than the student had been alive. Buy a Windows laptop for the stats work. Do it before the semester starts. The student wrote it down and did not buy the laptop. Six years later, the student was still doing the stats work, still on a Mac, and still asked by younger students whether the advice was true. It was true once. That is a different sentence than saying it is true now.
An advisor's advice is a record of the field the advisor learned inside. For a long stretch, that record about statistical software was accurate reporting. Serious quantitative work meant a Windows lab or a Windows laptop. The most-used packages were built for that lab. Cross-platform releases arrived later, ran slower on Mac than on Windows, and forced the researcher to accept a translated experience rather than a native one. The advisor was not showing preference. The advisor was telling the truth about the tools in front of the students.
That truth did not update on the same clock the platform updated on. The gap between the advice and the machine is where new researchers are getting stuck.
What actually changed under the Mac
The Mac of 2026 is not the Mac the advice was written for. Apple Silicon rebuilt the price-performance curve on the desk, in the backpack, and inside the fanless laptop the graduate student is actually using. Swift and SwiftUI made it possible to build a serious desktop application native to macOS, from the first line of code, without the compromises the older cross-platform frameworks forced. The open-source stats stack the field already trusts moved onto native arm64 builds in step with the platform. Front ends to that stack, built for people who prefer menus and dataframes over syntax, moved with them.
Say that plainly and the picture reorganizes. A Mac-first researcher in 2026 can open a coding environment that launches immediately, holds up under a long day of transcript work, follows the conventions of the operating system, and does not force them back through a system built for a different decade to get an answer.
The advisor's advice was written before any of that shipped. Nothing in the advice was dishonest. The platform under the advice moved.
The honest signal from inside the category
There is a moment inside the older vendors' own documentation that quietly resolves the argument. One of the category leaders, when a Mac user asks how to install its flagship statistical package, now sends that user to a different product from the same company. Not a workaround. Not a virtual machine. A different product entirely, built native, sold as the answer for Mac. The vendor is not fighting the point. The vendor is confirming it.
A researcher does not need a rival's opinion piece to hear that. The evidence sits in the vendor's own support docs.
The day the researcher actually has
The day the advisor imagined was a day in a Windows lab, next to a printer, on a machine chosen by the department for its ability to run the tools the department already owned. What the current researcher actually has is different. Transcripts open on a MacBook at a kitchen table. Coding happens in the same eight hours the writing, the meetings, and the childcare share. Battery lasts through the day. The researcher carries the same machine home and opens it again on the kitchen table. Judgment lands not on a feature comparison chart but on how well the tool holds up at eleven at night, three hundred excerpts in, when the deadline is real.
A tool built for that day is a different tool than a tool built for the day the advice was written for.
The tool the researcher lives inside all day is not a preference. It is a working condition.
What the update sounds like
The correct update to the advice does not embarrass the advisor. The advisor was accurate for an earlier era, and the field simply looks different now. Buy the laptop you already trust. Use the tools built for the machine you are actually going to open every day. The stats stack the field trusts is on that machine now, in a native build, running fast. On the qualitative side, longer neglected, native tools are catching up in step. Only a narrow lane of legacy enterprise software still holds to the older world, and the graduate student is not going to touch that lane for a decade anyway.
The advice made sense inside its era. Repeating it now, without updating it, hands the new researcher a working condition the platform no longer requires.
For the student in the first meeting
If a professor's advice this fall is to buy a Windows laptop for the stats work, the honest response is not to argue. It is to ask when the advice was last checked against the machines the field is now using. The answer to that question, in most rooms, will be honest and it will be dated.
The researcher does not owe the advisor an argument. The researcher owes the work a tool that respects the day it takes to do. Today's Mac earns that tool. Advisors can update what they tell new students this fall.
ReliCheck Quala is a native macOS qualitative analysis application built in Swift for Apple Silicon. ReliCheck Quanta is the same idea for the statistical side. See both at relicheck.com.