Persuasive Communication - How Audiences Decide. 2nd Edition

Cognitive Processes in Audience Decision Making 127

to invest in a fi rm (Firm A), the investment offi cer fi rst compared Firm A’s current value for an

important attribute to the values of the other fi rms in the client’s portfolio for the same attribute.

Next he compared Firm A’s value for that attribute to the values of other fi rms in the industry as

well as to mean growth rates and expected growth rates for that attribute. Then the investment

offi cer compared Firm A’s current value for a second important attribute to the values of the other

fi rms in the client’s portfolio, and so on.^204

Other expert audiences also show a preference for making attribute-based searches.^205 In a study

of expert auditors, seven senior auditors thought aloud as they planned an audit of a company’s

internal control procedures. The auditors were presented with the company’s fi nancial statements,

background information about the company, and fl ow charts of the company’s internal control

procedures. Most of the auditors’ time was spent making one of two types of attribute-based

comparisons: (1) the auditors compared two line-item amounts across years to determine if there

was a signifi cant change from one year to the next; and (2) they compared the information in the

company’s fi nancial statements to their conceptions of the information a prototypical fi rm would

disclose in its fi nancial statements.^206

In the case of executive selection, one study fi nds that higher level executives fi rst activate “their

schema of what it takes to perform well in a particular position.”^207 They then conduct an attri-

bute-based search comparing each candidate to the ideal candidate one attribute at a time. When

the comparison process is fi nished, the executives select the candidate who best matches the ideal.

Most searches conducted by consumers as they make decisions are attribute based. A think-aloud

study of consumer decision making fi nds that out of 80 searches, 57 (or about 71%) were by attri-

bute, 15 by alternative or brand, and 8 used some other search pattern.^208 A subsequent study agrees

that consumers use attribute-based search 71% of the time.^209 Another study that tracked consum-

ers’ eye movements as they decided among six used cars (alternatives) that were described in terms

of three decision criteria (attributes) also fi nds that consumers rely most heavily on attribute-based

search.^210 Consumers’ eye fi xations consisted of comparing pairs of alternatives one attribute at a

time. Even when consumers search by asking questions instead of reading about various products,

they typically use attribute-based search, with approximately 75% of their questions being attribute

based.^211

Audiences use attribute-based search in a variety of circumstances. In fi eld studies of consumer

decision making, attribute-based search strategies are the ones most frequently observed.^212 When

either alternative-based or attribute-based strategies can be used, audiences overwhelming use

attribute-based strategies.^213 Under time pressure, audiences’ search patterns become increasingly

attribute based.^214 Audiences’ search patterns also become more attribute based as the amount of

information about the alternatives increases^215 and when they have to trade off one attribute value

against another.^216

Even when audiences initially read and search by alternative, they may have to search their notes

or their memories a second time by attribute in order to make the comparisons their decision

requires. In a study of new home buyers, the buyers thought aloud as they decided among seven

houses that were described in seven separate booklets. In their initial reading of the booklets, buyers

conducted alternative-based searches. They read all about one house before reading anything about

another. But they conducted more attribute-based searches when reviewing the information in the

booklets a second time.^217 An eye-tracking study of judges choosing among scholarship applicants

who were described by three attributes came to a similar conclusion. Although the judges read the

information by alternative, they processed the information by attribute. Even when information

was structured to bias processing by alternative, half of the judges still processed that information

by attribute.^218