Why Students Drop Out in the First Term: Ask Before the First Assessment

Short answer

Students rarely decide to leave a course in a single moment. They fall behind, do not know who to ask, stop attending one session, then another. By the time withdrawal paperwork arrives, the decision is weeks old.

End-of-term surveys and withdrawal forms come too late and say too little. Ask a small group of current students, early in the first term, to describe the last time they got stuck and what they did next. Ask a few who left, if they agreed to be contacted. The answers show where support is missing while there is still time to add it.

Running a course or programme? Ask your own students.

Why course surveys and withdrawal forms miss it

Colleges, universities and training providers already collect a lot of student feedback. Most of it has the same three limits:

• Timing. Module evaluations arrive at the end. Students who left are no longer there to answer.
• Format. "I feel supported on my course: 3.1 out of 5" gives no clue about what kind of support, from whom, or when.
• The official reason. A withdrawal form offers boxes such as "personal reasons", "financial" or "course not as expected". Students tick the one that closes the conversation fastest.

Attendance and submission data add early warning. They show who is drifting. They do not show why, and "why" decides whether the right response is a tutor, a timetable change or a hardship fund.

The first weeks matter most

Many of the moments that shape a student's decision happen before the first graded work:

• the first session they do not understand;
• the first time they miss a class and are not sure how to catch up;
• the first group task with people they do not know;
• the first clash between a shift at work and a lecture.

How each of those goes depends on small things: whether they know who to email, whether anyone noticed, whether asking for help felt normal. Those are exactly the details a conversation can collect and a scale cannot.

What to ask students

Ask the last question plainly. Students who have thought about leaving are usually willing to say so when the question is put without judgement.

Who to invite, and when

• Ten to fifteen first-year students in weeks three to five, from different groups and backgrounds.
• A few students whose attendance has dropped, invited the same way as everyone else.
• Students who withdrew last year, if they agreed to be contacted.
• Two or three second-years who nearly left and stayed. What kept them is evidence.

Tell students who will read their answers and that taking part has no effect on their marks. Keep tutors who grade them out of the reading if you can.

A fictional example: a first-year cohort

A fictional college loses about one in five students from a first-year business course before the second term. The module survey shows "I feel supported on my course" at 3.1 out of 5. The course team plans extra revision sessions before the first assessment.

In week four, the programme lead asks ten students one question: "What has made the course harder than you expected, and what did you do about it?"

• Seven of the ten say they did not know who to ask when they fell behind. Several name the wrong person.
• Six say they first felt lost in week three, during a statistics unit, well before any assessment.
• Four work evening shifts that clash with the only drop-in help session.
• Three have thought about leaving. All three are among the seven.

Revision sessions before the assessment would arrive after the point where students say they lost the thread. The conversations suggest earlier, smaller steps: a named person to contact, introduced in week one, and a second drop-in session at lunchtime.

Ten invented students prove nothing about the whole cohort. They do move the team's attention from week eight to week three.

What to do with what you hear

1. Group by moment, such as "week three statistics" or "missed session", rather than by student.
2. Count how many students described each one.
3. Pick changes that can happen this term. A named contact and a second help slot are quick. A curriculum rewrite is slow.
4. Tell the students what you changed.
5. Ask the next cohort the same question in the same week.

Care with student conversations

• Some answers will touch on money, health or family. Say at the start what the study is for and where students can get support.
• Do not promise that answers cannot be linked to a person. In a small cohort, a detail can identify someone.
• If students are under 18, follow your institution's safeguarding and consent rules before inviting anyone.
• Report conditions and moments. Leave out who said what about which tutor.

Where Lontra fits

With Lontra, you write the question in one sentence and review the topics it proposes. Each student gets a private link, answers in writing on a phone with no account, at a time that suits them. Lontra follows up on what they say, for example asking what happened after they got stuck. You read findings with how many students described each point, where they disagree, and the quotes behind them. It is free to try with 10 conversations. It is a conversational AI survey: one open question and follow-ups, in place of a rating form.

It is not a welfare service and does not replace a tutor noticing that someone has stopped coming. It does not give a retention statistic. Participants are told before they start that you will read their answers. See who sees what.

Ask in week four

Withdrawal data explains last year. A question asked in week four can still change this one. Start with your question and invite ten students.