Proximity: Localization vs. Distance in PR News Releases

Charting the use made of 197 news releases mailed from a Mississippi university to 121 daily and weekly state newspapers tested this question: which results in more use, the proximity of the public relations release source or localization of the facts in the story. The influence of proximity proved to be very slight but localization of stories resulted in relatively higher use. This study compared costs and concludes that localization of public relations news releases may be worth — in the age of word processing computers — the small extra expense.

pul)lication decisions of 36 gatekeepers with two of five types of editors selecting stories primarily on proximity.2 Aronoff also used proximity to mean both geographic distance and localization and found it to correlate "most highly with the acceptlreject variable."3 Similarly, Clyde a n d Buckalew considered both geographic distance a n d localization and found that editors ranked news items having the news elements conflict, proximity and timeliness higher than those without them, and higher than those with the news elements impact or known principals.4 Several researchers w h o isolated geographic d i s t a n c e from Ii.)calization found it not to be as good a predictor of publication as when the two c o m p~n e n t s were considered together. Luttberg studied a plane crash, a major event for the 75 newspapers located close by. He found that those "papers covering a story on average" were only "two miles closer than those omitting coverage." He concluded that "actual distances t i i each day's events play little rnle in their (gatekeepers') "... judgment as to which stories to include in their paper.5 Martin hypothesized that newspapers closest to a community where an event takes place publish more stories about it, but his findings caused him to note a possible confounding of the proximity element: The Corin'er-jorrmnl and Times, though sister publications of the news organization, gave different amounts of space to the contract issue from one another. This somewhat confounds the results, suggesting a factor other than simple organization proximity.6 Shoemaker and Mayfield also interpreted proximity as distance and c:oncludetl that it "may not serve as a good measiirt! of newsworthiness .... ''7 Other studies looked specifically at how localization affected the ;rcc:el)tmce of news releases. Abbott and Brassfield defined proximity as "the localness factor" and found it to be "the single most important factor for both tt:levision and newspapers in accepting or rejecting a news release." The gatekeepers in their study preferred news releases which were tailored to the locality ant^ nietIiuni.8 Several other studies have found a lack of localization or local interest to be a frequent reason that editors cite for rejecting releases.9 One by Elfenbein surveyed 170 business editors nationwide to determine their evaluations of news releayes. They stressed their need for hometown angles or localized information. Yet the editors in Elfenbein's study reported that most (95%) of the releases they received did not refer to "local companies, industries or interests." Consequently these releases were not used.1° Likewise, Turk found little attempt to localize information. On the contrary, the state agency public information officers in her study gave the "same information news release ... to all newspapers, whether there was a specific local angle nr nnt.**11 Hypothesis 1: Placement of general news releases is negatively correlated with the geographic distance Ijetween the newspapers and location where the release was issued.
Hypothesis 2: Placement of localized releases is negatively correlated with the geographic tlistance IJetween the newspapers and location where the release was i s s l i d Hypothesis 3: Placement of localized releases is greater than that of general releases regardless of geographic distance between the newspapers and location where the release was i s s u d .
-The subjects of this study are 197 press releases niailed from the public relations tlepartment of a regional Mississippi university (luring the six-month period from Jone to Deceniber 1988. Of the 197 releases, 23 were general in nature, with no attempt to localize their contents. A total of 1,774 copies of these 23 releases was distributed to Mississippi newspapers. The 174 other releases were localized by being partially rewritten for each newspaper to which they were sent.
A localized release was defined as one in a series of releases in which the lead and associated niaterial hail been rewritten to be of interesi to the specific paper to which it was being sent. For example, a release concerning the awarding of st:hol;irshiIis might contain a list of the recipients. General releases would have a generic lead with the list of recipients in the body of the text. In a localized release, the lead woriltl contain the name or names of intlivitluals resitling i n t lie circulation area of the paper to which that release  w a s being sent. Because totally custom-written releases, s u c h as honietowners, might bias the study in favor of the hypotheses, each release iisetl in the study had to contain at least 55% of text common to the series. Percentages were calculated by measuring the release and the common material. If any paragraph contained information unique to the release, the entire paragraph was considered localized. The releases were mailed to 121 daily and weekly newspapers in

Mississippi. The Conrmerciol D i s p n t c h , a newspaper in Columbus.
Mississippi, was excluded from the study because it was within two miles of t h i ! university, antl therefore, all releases from that source would be t:onsiileretl local. Because geographic distance as well as localization has liean found to influence piiblication, geographic distance was used as a covsriate. The actual mileage of each paper from the university was used to test hypotheses one and two with Pearson Correlations. For hypothesis three the distance was coded on a five-point ordinal scale: 1 =<50 miles, 2 = 50 to 99 nules, 3 = 100 to 149 miles, 4 = 150 to 199 miles, 5 = >lo9 miles. A 2x5 niixetl design M A N O V A was used with repeated measires for the percent of general antl localized releases placed as the within factor and geographic tlistance as the between factor. A significance level of .01 was used.
A professional clipping service was used to monitor use of the press releases by the newspapers. A release was considered used if it was puldis1it:tl substantially in its original form.
Of the 1,774 copies of the 23 general releases, 87 (5%) were used by 39 newspapers. Of the newspapers that used general releases. 35% used 10% or fewer of the releases. Of the 174 localized releases, 78 (45%) were used. Eighty one newspapers received the localized releases. Of this number, 47 (58%)) used at least one release. and more than 17% used every release sent t 111: 111.
Findings on Hypotheses 1 and 2 No significant correlation was found between newspapers' acceptance rates for either general or localized releases and geographic distance (r for g e n d releases = -.1327, 1) =.074; r for localized = -.1577, p =.O43). Although the correlation for distance with localized releases was in the predicted tlirection and woiiltl have been significant if the common significance level n l .05 had been used. the small coefficient indicates only a slight rc:lationship. The usage rates of the general and the localized releases were tested with a 2 X 5 repeated-measures MANOVA with distance as a covariate. Results are shown in Table 1. Localized releases showed a much greater usage rate than general releases. This study indicates that the positive relationship between proxinuty and publication is actually a result of localization. Although there was some twidence that newspapers closer to the source showed a slightly greater tlisposition toward Iiihlisliing a general release. this disposition was slight. However, as means in Table 2 indicate. localized releases were significantly more likely to be piiblished than a general release regardless of the recipient newspaper's distance from the soiuce.
This study confirms with controlled, statistical evidence previous research on the positive effccts of localization. Although practitioners have lor some time acknowledged a vague belief in the advantages of localization, studies like Turk's indicate that they do not localize releases.12 Several explanations of this phenomenon snggest themselves. First, the evidence nxiy have been seen as sufficiently weak so that practitioners rejected it. However, this study indicates that the effect is far from mild.
Second, practitioners may have experienced the inertia that manifests itself when a change requires atltlitional work. However, in the present study, the 174 local releases resulted in almost as many publications (78) as the 1,774 copies of the gmeral releases (87). With today's standard use of computerized word-processing equipment, localizing releases is not nearly as much work as in the past. For instance, in timing the difference between production of a generalized release and five localized releases, the difference from composing through copying was only ten minutes or two minutes per loralized article. Since mailing antl copying costs are the same per release, this time difference ;it sin per hour equals a total cost difference of only 33.3 c m t s per localized release. Furthermore. it could easily be argued that preparing antl mailing 80 localized releases is less trouble than photocopying antl mailing 1,000 copies of general releases.
Third. accoiintal)ility on the part of public relations professionals has often been definetl R S nunitier of' releases sent out rather than the number placed. However, as piiblic relations matures, the emphasis is shifting from volume to effect. The accepthject ratio of releases is one criterion upon which campaigns are now being measured. Localization allows the practitioner to improve effects with very little extra effort. For instance, of the releases in this study. less than a quarter (in = 22.12%) of the contents were localized, ant1 77% had less than 30% localized material.
Localizing releases may improve practitioners media relations and improve credibility. Even though research shows that journalists have a better relationship with PR professionals whom they deal with f r e q~i e n t l y ,~~ it's reasonable to postulate that credibility of a source from which journalists ;ire constantly receiving unusable information decreases. Logically, the converse would apply when releases are localized and, as such, more usable. (This relationship between localizing releases and increasing credibility needs to be studied.) Fortunately for public-relations practitioners, geographic distance, over which they have little control, seems to have little effort over editors' tlecision-making while localization, which they can w e as a tool, seems to be H significant lactor.