Tony Ford

Eyewitness Identification Research Laboratory
At the University of Texas at El Paso -

Their study and report:

Tony Ford is residing on Death Row in Texas awaiting execution. The 5th Circuit Court of Appeals denied his appeal. One of the issues in the appeal was denial of funds to hire Professor Malpass to testify as an expert concerning the eyewitness identification issues in the case.

Mr. Fords' appelate attorneys asked Professor Malpass and the Eyewitness Lab to evaluate the lineup used in the investigation that resulted in his identification, and to do certain other studies concerning the identification issues. We began with an evaluation of the lineup, based on the principles and procedures described in previous pages on this section of the Labs website.

The witnesses in this case produced extremely impoverished verbal descriptions of the perpetrator.  It is not possible to determine whether this results from their having gotten only poor visual exposure to the perpetrator during the event, thus having little information on which to base a description, or whether it resulted from a lack of attention to getting a verbal description on the part of law enforcement’s first responders, and subsequently by the investigators.

Since no useful verbal descriptions were available in the record, we elicited verbal descriptions from 20 UTEP students recruited as participants specifically for this purpose. We showed these participants the lineup photo of Mr. Ford and gave them this instruction:

“Please write a description of this person in enough detail so that someone else could identify him on the basis of the description.”

From the resulting descriptions we compiled the following composite description, using the most frequently mentioned descriptor provided in each of the facial feature categories:

African American male in his early 20’s with dark hair.
An oval face and broad forehead. 
Small, dark eyes and thin eyebrows. 
A wide nose, thick lips and small protruding ears.

 Lineup Evaluation Procedure

A color copy of the lineup used by the El Paso Police Department was provided to us.  This copy was scanned into our computer and used in this evaluation.

We displayed the composite verbal description to 40 new Hispanic participants of college age, and subsequently asked them to examine the photospread and identify the police suspect. These mock witnesses were each seated individually in front of a computer screen.  The following instructions were displayed to them:

First Screen.

"Thank you for agreeing to participate in this experiment.  We will explain the purpose of the study once we have finished.  We are going to ask you to read a description of a person.  Please read this description carefully.  You will be asked to choose the person who best matches the description."

Second Screen.

"Person description: African American male in his early 20’s with dark hair, an oval face and broad forehead.  Small, dark eyes and thin eyebrows.  A wide nose, thick lips and small, protruding ears."

Third Screen

"In a moment you will be asked to view a photographic lineup.  Please choose the lineup member that best matches the description you just read."

Following this instruction the lineup was displayed, and participants used the computer mouse to click on the lineup position number of their choice.

The lineup is displayed on the following page. See if you can determine which of the 6 individuals is Mr. Ford.

Thirty nine Hispanic people from the UTEP campus examined the lineup and made a choice. The expected ID rate per lineup member is 6.5. Here are the results:

 

Mr. Ford is in position #5. Obviously the lineup is biased towards picking Mr. Ford, because he stands out from the fillers in the lineup. The expected identification rate in a six-person lineup is 1 / 6 = 16.7%. Twenty two choices out of 39 is 56.4%. We are more than 99% certain that the rate at which Mr. Ford was chosen exceeds the level of chance expectation (16.67%), and are very confident that the observed rate of identification is larger than would be expected if only chance factors determined participants lineup choices.  It is likely that some systematic process is occurring to direct choices to Mr. Ford.

 

It is not difficult to discern what these might be. Note that the 3 lowest frequency lineup members taken together (a total of 5 identifications) do not add up to the number of choices expected for one fully adequate lineup member. As safeguards against false identification these three fillers may as well not be there, and the risk of false identification for an innocent suspect is about 1 in 3, or 33%. The 3 ineffective fillers in this lineup fail in their role as a safeguard against false identification. An additional filler fulfills only 66% of its expected performance as a safeguard against false identification of an innocent suspect.

 

It is not difficult to understand why these fillers are inadequate when you examine their degree of resemblance to the suspect in this lineup - Mr. Ford.

 

Mr. Ford's Appelate Attorneys felt that another suspect in the case (Victor Belton) looked remarkably like Mr. Ford, and thought it possible that had he appeared in a lineup instead of Mr. Ford that he might have been identified by the eyewitness. Therefore they asked us to carry out a study of the similarity of Ford and Belton in the context of lineup fillers. (See Burr, 2005)

 

A new set of 80 participants were used to provide ratings of the similarity between the photos of Ford and Belton to each other, and to each of a) the five fillers from the Ford lineup, and b) five additional fillers we selected from the same mug-shot database used in the Ford lineup.

 

We presented participants with the 21 pairs of faces shown in the Table below, in random order.  Participants were asked to rate the similarity of the two faces in each pair on a scale ranging from 1 to 7 (from “not similar” to “very similar”).

The purpose of the study was to determine the degree to which Ford and Belton are similar in appearance, and whether it is possible that they could be mistaken for each other.  Our results show that of all the 21 face photo pairs in this study, the Ford and Belton pair is rated as the most similar, and significantly more similar than the next most similar pair. Belton is more similar to Ford than any of the fillers used in the Ford lineup, in many cases by a considerable margin.

Mean Similarity Ratings, pairing Ford and Belton
with each other and with each of 10 lineup fillers: