Source: https://akholi.com/outsourcing-2026-agentic-ai-tier-1-banks/study-data/part-4-analysis-experience-by-intent-to-leave/
Author: Phil Hatch
Publisher: Akholi

Years of experience by reported intent to leave: C007 by C019

Study Data · Part 4 · Indicator Deep-Dive Analysis

# Years of experience by reported intent to leave: C007 by C019

Indicator ANALYSIS:experience_by_intent_to_leave · By Phil Hatch · Akholi · October 9, 2026

← Study Data← Previous indicator

## On this page

20 sectionsIndicator SpecificationSummary of FindingsAnalysis ResultsDistribution of ResponsesDescriptive Statistics by SegmentTests of Difference Between SegmentsAssociation with Cancellation RateAssociation with Customer SatisfactionAssociation with the Engagement Indicator MeanAssociation with Intent to LeaveComparison of Associations Across the Four MeasuresClustering and Effective Sample SizeEquivalence Tests of Null ResultsRobustness ChecksMeasure Independence and Part-Whole CorrectionComparison with the Study NormInterpretation of the DataLimitationsDefinitionsSources and Provenance

## Indicator Specification

Table 1. Indicator specificationTable

| Field | Value |
| --- | --- |
| Indicator code | None, a derived analysis |
| Question as asked | Years of experience by reported intent to leave: C007 by C019 |
| Failure category | Not one of the 100 rated indicators |
| Place in the survey | Not a survey item |
| Indicator class | Derived from other indicators |
| Indicator type | Derived analysis |
| Response scale | Not applicable, a derived analysis |
| Answering base | As stated in Analysis Results |
| Non-answers | Not applicable |

## Summary of Findings

Respondents likely to leave their job, those who answered C019 at seven or above, reported fewer years of career experience (C007) than respondents not likely to leave: a mean of 3.77 years (n = 409 respondents) against 4.44 years (n = 337 respondents). The difference, −0.66 years, reached significance by Welch's test (t = −2.04, p = 0.041) and with standard errors clustered by engagement (t = −2.64, df = 149, p = 0.009, 150 engagements). The effect is small (Cohen's d = −0.15). Across the 746 respondents with both answers, years of experience correlated with intent to leave at r = −0.09 (p = 0.002, clustered by engagement).

## Analysis Results

The first table gives the base, the rule and the two groups. The second gives the tests of the difference and of the correlation.

Table 2. Analysis resultsTable

| Result | Value |
| --- | --- |
| Unit of analysis | Respondent |
| Respondents with both answers | 746 |
| Likely to leave, respondents | 409 |
| Likely to leave, mean years (SD) | 3.77 (4.16) |
| Likely to leave, median and range | 2.0, 0.0 to 20.0 |
| Not likely to leave, respondents | 337 |
| Not likely to leave, mean years (SD) | 4.44 (4.62) |
| Not likely to leave, median and range | 3.0, 0.0 to 25.0 |

Table 3. Tests of the difference and of the correlationTable

| Test | Estimate | Standard error | t | df | p | n | Engagements |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Difference in mean years, Welch's test | −0.66 | | −2.04 | | 0.041 | 746 | |
| Effect size, Cohen's d | −0.15 | | | | | | |
| Difference, clustered by engagement, as r | −0.08 | 0.03 | −2.64 | 149 | 0.009 | 746 | 150 |
| Years with C019, clustered by engagement, r | −0.09 | 0.03 | −3.13 | 149 | 0.002 | 746 | 150 |

## Distribution of Responses

The distribution of years of experience is reported in the file of C007, and the distribution of intent to leave in the file of C019. In both groups the median lay below the mean (2.0 against 3.77 years among respondents likely to leave, 3.0 against 4.44 among the others), so a minority with long careers raised each mean.

## Descriptive Statistics by Segment

The analysis compared the two groups as a whole and was not broken down by segment.

## Tests of Difference Between Segments

No test of difference between segments was computed for this analysis. The only comparison is between respondents likely and not likely to leave.

## Association with Cancellation Rate

Cancellation was not part of this analysis.

## Association with Customer Satisfaction

Customer satisfaction was not part of this analysis.

## Association with the Engagement Indicator Mean

The engagement indicator mean was not part of this analysis.

## Association with Intent to Leave

Intent to leave is the answer to C019, “How likely are you to leave your current job in the next 12 months?”, rated 1 to 10. The analysis splits respondents at seven or above, the study's definition of likely to leave, and also correlates the full rating with years of experience. Both forms gave the same direction: fewer years of experience came with higher intent to leave.

## Comparison of Associations Across the Four Measures

The analysis concerns intent to leave alone, so no comparison across the four measures applies.

## Clustering and Effective Sample Size

Respondents on the same engagement share conditions, so the tests were repeated with standard errors clustered by engagement across 150 engagements. The clustered test of the difference gave p = 0.009 and the clustered correlation p = 0.002.

## Equivalence Tests of Null Results

No equivalence test was computed, because both tests found a difference from zero.

## Robustness Checks

The difference was tested two ways, by Welch's test treating respondents as independent and by a regression with standard errors clustered by engagement. Both reached significance. No other robustness checks were computed.

## Measure Independence and Part-Whole Correction

Years of experience and intent to leave are separate questions, and neither enters the other. No part-whole correction applies.

## Comparison with the Study Norm

A derived analysis has no profile against the study norm, so no comparison applies.

## Interpretation of the Data

Respondents likely to leave had somewhat shorter careers than the others, a difference of under a year on means of about four years, and the association, though it held when clustered by engagement, is small. Years of experience track respondent role closely, as the file of C007 reports, so the difference cannot be separated from role here.

## Limitations

The evidence is the reported experience of the vendor's delivery teams, collected in a single cross-section. The association reported here is not causal.

Years of experience are the respondent's report. A report of zero means under one year, so the shortest careers are all recorded as zero.

## Definitions

Table 4. DefinitionsTable

| Term | Definition |
| --- | --- |
| Likely to leave | A respondent who answered C019 at seven or above. |
| Years of experience | The answer to C007, years of career experience, where zero means under one year. |
| Cohen's d | The difference in means divided by the pooled standard deviation. |
| Clustered by engagement | A test whose standard error allows for the answers of respondents on the same engagement to be alike. |

## Sources and Provenance

Table 5. Sources and provenanceTable

| Source | Value |
| --- | --- |
| Study data | Study Data.xlsx, sha256 219eba0457274d04 |
| Statistics engine | compute_statistics.py, sha256 738319eef764602b |
| Statistics record | Computed Statistics.json, record ANALYSIS:experience_by_intent_to_leave, sha256 1ae50a3637fc9370 |
| Computed | October 13, 2026, 04:57 UTC |
| Template | Indicator Template v2, version 2.0 of 10 October 2026 |
| Figures placed by script | 59 |