After a fortnight of what Niki Savva of The Australian described as “an opportunity to see Tony Abbott stripped of politics”, the verdict of Australia’s poll respondents is in. Five polls conducted entirely after the MH17 disaster of July 17 find the electorate to be more moderate in its hostility towards Tony Abbott, and to have become slightly less ill-disposed towards voting for his government.

This less-than-spectacular finding is based on 2-point shifts in the Coalition’s favour in polls from Essential Research and Morgan, and 1-point shifts from Galaxy and ReachTEL. However, the change seems to have got lost inside the error margin at Newspoll, which this week had Labor’s lead unchanged at 54-46.

The effect was more pronounced on personal ratings, with even Newspoll recording a 12% improvement on Abbott’s hitherto dismal net approval, up from from minus 29% to minus 17%, and finding him level with Bill Shorten as preferred prime minister for the first time in three months. All of which adds up to noticeable movement to the Coalition in the BludgerTrack poll aggregate, which now has Labor’s two-party lead of 52.1-47.9.

Modest though the shift may be, it’s enough to raise the question of whether a grudging but durable respect has developed for Abbott over the past fortnight, or if the public has merely relaxed its prevailing cynicism for a time in response to a national tragedy.

Some perspective on this might be offered by looking at the past record of leaders who experienced improvement in their poll ratings after major disasters. Five that spring to mind are the Port Arthur shootings of April 1996, the September 11 attacks of 2001, the Bali bombings of October 2002 and, at state government level, the Victorian bushfires of February 2009 and the Queensland floods of January 2011.

Trend readings of opinion poll data from each period are displayed on the charts below, with the nascent recovery of the Abbott government shown in black for comparison.

The two charts show the two-party vote for the governing party and the net approval rating (i.e. approval minus disapproval) for the prime minister or premier over a period from 30 days before the date identified as the peak of the event in question, to 60 days thereafter.

The results suggest that Abbott is following the typical pattern in enjoying surging personal ratings but a mediocre yield on voting intention. Such was the case after Port Arthur, when John Howard — who was already at a honeymoon high less than two months after coming to office, with apparently little room left to improve — gained 8 points on net approval while voting intention remained unchanged.

The September 11 effect is likewise complicated by the Tampa episode having unfolded a fortnight previously, with the two events seeming at the time to reinforce each other in boosting Howard’s stocks. Clearly the attacks accelerated the decisive shift to the Coalition already under way, but it’s an under-discussed fact of the 2001 campaign that the surge dissipated markedly when the election campaign reasserted the salience of domestic issues.

Even so, August/September 2001 emerges as a clear example of crisis leaving a leader durably strengthened, to the extent that Howard’s strong personal ratings remained intact when the Bali bombing occurred a year later. A further 10-point gain after the bombing proved temporary, but a small lead that opened up on voting intention was still in evidence three months later, as Labor struggled to land a glove on the government under Simon Crean’s leadership.

Polling from early 2009 shows why John Brumby’s response to the February bushfires in Victoria was seen at the time to have strengthened his grip on power 18 months after he took over from Steve Bracks, in what should be seen as a cautionary tale for those who would prognosticate over poll trends at long range.

It’s also interesting to observe that Anna Bligh was the only leader with comparably poor ratings to Tony Abbott a month before the onset of their respective crises. The poll surge to Bligh was several orders of magnitude greater than the one Abbott is clearly going to experience, for all the good it did her in the end (it should be noted that the results for Queensland in the charts are based on a limited number of polls, so what was in fact a sudden shift after the floods shows up as a gradual movement over two months).

It could be argued that the more auspicious examples of September 11 and the Bali bombing offer better points of comparison for MH17 than a natural disaster, owing to their shared foreign policy and national security dimensions. But even so, the story that most forcefully emerges from the charts is that Abbott is still deep in the doldrums even after a fortnight when the media agenda has been playing to his strengths, and at a time in his young government’s life when it ought to be ramming home the advantage over a defeated and demoralised opposition.