The Effect On Ratings Linked To Moving Programs Within
The Prime Time Broadcast Schedule
By William Jenson Adams
A.Q. Miller School of Journalism
And
Mass Communications
105 Kedzie Hall
Kansas State University
Manhattan, KS 66505
[log in to unmask]
The Effect On Ratings Linked To Moving Programs Within
The Prime Time Broadcast Schedule
This study looked at the relationship between moving programs and the ratings.
While moving an established series once during the main broadcast year improved
the chances of a significant rating increase for both the series and the
network, moving an established series during the summer resulted in significant
rating decreases. Whether or not a move benefitted a new program depended on
how the new program was doing in its original slot. Neither new nor established
series survived more than one move in any given year.
The Effect On Ratings Linked To Moving Programs Within
The Prime Time Broadcast Schedule
The fact that the prime time schedule on the broadcast networks has become less
and less stable over the last two decades is indisputable. The number of
programs moved around within the schedule, and the number of programs replaced
in less than 6 weeks between September and April went from 44 in 1971 to over
130 by 1997. The pattern for these shifts in the schedule also changed
dramatically. In the early 1970s, the vast majority of moves occurred in either
September (when the new season started) or January (when the second season
started), and cancellations in less than one season were virtually unheard of.
However, by the late 1990s, shifts were occurring constantly, with the number of
changes in October, March and April all but equaling those that occurred in
September and January (Adams & Eastman, 1997).
These changes now involved more than simply replacing failed new series, as was
the case in the early 1970s. Now, many programs simply trade places, or are
shifted to new times or days seemingly just to see what will happen. In the
1997-98 season alone for example, 24 prime-time programs were moved from one
time slot or day to another, and 11 were moved more than once.[1] The vast
majority of these moves were not done to replace a failed series, but instead
replaced another moved series or a program that was taken off the air for a few
weeks then returned to the same time slot. With all of this activity, one would
think there would be a great deal of research to determine what the effect of
this instability was on the ratings. That thinking would be wrong.
A review of the academic and popular literature in the broadcast field reveals
virtually no scientific research has been published regarding the effects of
programming instability. This may be a result of the fact that instability is
not a programming strategies as such, but rather, a means to achieve other
programming goals. For example, programs are canceled and replaced quickly in
order to sure up the ratings in weak time slots, while series are moved to
create strong lead-ins, programing blocks,
Rating Change Linked To Program Moves
counter programming and so on. These programming strategies - i.e. blocking,
lead-in, counter programming, etc. have been studied a good deal during the last
decade (Adams, 1997; Eastman, Newton, Riggs & Neal-Lunsford, 1997; Cooper, 1996;
Lin, 1995; 9 Adams, W. J., 1993; Livingstone, 1993; Morley, 1993;
Brosius, H., Wober, M. & Weimann, G., 1992; Boemer, M. L., 1987; Tiedge, J. T. &
Ksobiech, K. J., 1987; Wakshlag, J. & Greenberg, B. S., 1979 ) with much of the
research suggesting their effectiveness may be limited. However, the effects on
ratings, or lack there of, connected directly to program instability remains a
mystery.
This is not to say the constant shifting of programs has not been discussed.
Todd Gitlin (1983), in his book Inside Prime Time, points out such instability
is the natural result of network programmers' efforts to create the "optimum"
schedule - - i.e. a schedule that will produce the best possible overall ratings
for a network. He went on to say the shifting of the schedule: "is no longer
an annual affair, but a year long process of rearrangement, like a frantic,
continuous round of interior redecoration." (p.60)
While no one disputes the fact that program instability is a basic part of
present day scheduling, some within the industry have questioned just how useful
it really is. Paul Klein, the man often credited with starting the phenomenon,
argued against moving programs. He claims that such moves will not change the
basic make-up of the audience at any given time. In short, people watch
televison based on availability, not programming. In his view, moving programs
around would have no real effect on the overall rating, but might anger viewers
(Klein, 1971).
CBS researcher Arnold Becker calls the continual schedule shifting "playing
with yourself." He claims proprietary research by CBS documented the fact that
all the "schedule-juggling", as he calls it, had no real effect on viewing, but
did cost a great deal of money. According to him, for every increase there was
an equivalent decrease some where else in the schedule. As a result, millions
of dollars were spent just to end up in the same rating position as before
(Gitlin, 1983). Becker goes on to suggest schedule juggling is an act of
desperation based on the fact that ratings have become all important. In his
view, programmers, faced with the fact that they can't improve the programs now
on the air, and, having nothing better to put in their place, are left with only
one option when it comes to improving the ratings. They continually "move
things around and pray (Gitlin, p.61)."
Deanne Barkley, former vice president for movies at both NBC and ABC, has a
more cynical view. She argues that programmers know the endless adjusting of
the schedule is ineffective. She claims the networks could actually do a much
better job of programming with far fewer people, and contends programmers know
this. Schedule juggling is how a lot of these extra people justify their
enormous salaries and existence (Gitlin, 1983).
Gitlin points out however that these people are in the minority. Most
programmers firmly believe that fine tunning the schedule is a necessary and
valuable tool. Even the critics mention above don't believe such practices are
harmful. They merely argue they are ineffective, unnecessary and expensive.
However, recent qualitative work suggests schedule instability may be one of the
major factors driving people away from the broadcast networks. Participants in
focus groups claimed the constant changing of the schedule caused frustration in
viewers and led them to seek else where for the stability they want. They also
said such schedule shuffling interfered with the mental images people have
regarding when desired programs are suppose to be on (Adams, 1999). Of course,
this study is based on perceptions and has no quantitative results to back it up
at this time. As a result, the question of what effect moving programs has on
the ratings is still wide open. This study will seek to provide some answers.
Specifically, it will seek to answer the question of what effect moving programs
within the broadcast networks' prime time schedule has on the ratings.
Method
The definition of what constitutes a moved program proved a little more
complicated that might be assumed. For example, do programs moved during the
summer break - i.e. series that run in one time slot through May and then return
in another slot in September - behave the same as series moved during the main
viewing year (September to May). Further more, do established programs behave
the same as new programs, and, more specifically, what constitutes an
established show? Traditionally, an established series has been considered a
show that was renewed for another year. However, beginning in the 1980s, the
major networks began running new series in March and April to fill time and save
original episodes of popular show for the May sweeps. These series would run
four to six weeks, then be canceled or renewed for another year. As a result,
if renewal is used as the definition of established, a series that ran 4 weeks
could be considered established while another series which ran 32 weeks would be
considered new. To solve these problems, programs were separated based on the
following definitions:
1. An established series was defined as a program that has run 32 weeks (two
full seasons) and then been renew for another season.
2. All programs within the first 32 weeks were classified as new. (Originally,
series run only in March or April and then renewed for another year were
classified as limited run. However, test found these series behaved no
differently overall than other new series, so this division was dropped.)
Movement of the programs was further separated into the following division:
1. Programs which were moved during the summer-i.e. series that ended their run
in one place in May then started their run in September in a different slot.
2. Programs moved during the main viewing seasons - i.e. between September and
May.
3. Programs moved more than once during any single year.
4. Programs where the move resulted in a return to a time slot the series had
previously occupied within the last 32 weeks. (It was reasoned that such a move
would allowed researchers to see if returning a series to its former time slot
would result in a return to its former rating.).
Determining a change in the ratings also required more than one measurement.
The study certainly wanted to look at the change in rating for the program that
was moved. However, the hypotheses surrounding optimal scheduling are less
concerned with individual programs than with the overall rating change for the
network. For example, say a popular series was moved and dropped in the ratings
from 20 to 16. On the surface that would seem to be major problem. However, if
the new program put in the old time slot held the 20 and the moved series
improved its new time slot by say 4 points, from a network point of view the
move would be a success. For that reason the following four rating changes were
recorded for each series moved:
1. The change in rating for the series itself.
2. The change in rating for the time slot it was moved out of.
3. The change in rating for the time slot it was moved into, and
4. The overall change in rating for the network.
The ratings were taken from both Variety and Broadcasting and Cable magazine.
The ratings were then averaged for the series in the specific time slot for that
specific year. To see if the strength of the show was an important factor, how
the program ranked before and after the move, the rating in the original time
slot and how that rating compared with the rating in that time slot for the
previous year was also recorded.
Established and new series that were not moved were used as the control.
Rating changes under the tested conditions were compared to rating changes under
the never moved category to determine any difference that might have occurred.
Each year was recorded separately, so a program could be counted more than once,
as most established programs were.
All programs offered by ABC, CBS, NBC, FOX, WB and UPN in prime time between
1986 and 1997 were recorded, making this a population study of programming
during this period. This resulted in 458 established series that were not move,
569 new series that were not move, 178 established series moved during the
summer, 45 new series moved during the summer, 137 established series moved
during the broadcast year and 190 new series moved during the broadcast year.
Because FOX, WB and UPN were used in this study, it was not always possible to
determine the change in rating for all original time slot. In eight cases, as
far as these three networks were concerned, the program moved was the first show
ever to occupy that slot making it impossible to say whether the series had
improved things or not. There were also a few cases where changes in the FOX
schedule made it impossible to determine the rating change for the original time
slot when the series was moved as the program was not replaced.
As this was a population study, statistical test were limited. However, as
ratings are based on sampling, statistically significant differences were
determined using Nielsen's tables. For this study, statistically significant
changes are any changes larger than one standard deviation. Two standard
deviations are much more common in academic research, but they proved
impractical for this research as rating difference during the period studied
tended to be small. Under the two standard deviation system, there was often no
significant difference between the tenth and fiftieth ranked series. For this
reason, broadcasting traditionally uses one standard deviation when discussing
rating changes. This study adopted that practice.
Results
The relationship between moving programs within the prime time schedule and
ratings change was different based on the type of move and the type of program
involved. The effects upon the program were also not always the same as the
effects upon the network. The results from the study of established series are
a good example.
Moving Established Programs
As can be seen on Table 1, moving an established series during the summer
increased the chance of a significant rating decline in all categories except
for the new time slot (the one the series was moved into). The original time
slot declined significantly 17.5 percent more often, while the series itself
declined 30 percent more often and the network declined 14.6 percent more often
than when the established series was left alone. The chance of a significant
decline in the new time slot remained virtually the same as when no move
occurred.
On the other hand, the new time slot increased its ratings significantly 28.7
percent of the time. As a result there was also a 7.7 percent increase in the
networks chance of increasing its overall rating significantly. The table makes
it clear how the network could both lose and gain from such a move. When an
established series was not moved during the summer, almost 50 percent of the
time there was no significant rating change from one year to the next. However,
when an established series was moved, this stability dropped dramatically. As a
result, a move meant a clear loss for the series, but a gamble for the network.
As the table shows, it was a gamble the networks lost twice for every time they
won.
Table 1
Comparison of Rating Changes For Established Series That Were Moved
To Established Series That Were Not Moved
Established Established Series Established Series
Not Moved Move
Moved During Summer During Year
Number of Programs 458 series 178 series 137 series
Original New Series Net. Original New Series
Net.
Increased Rating By
At Least 1 STD. 11.4 14.5 28.7 12.4 19.1 16.1
33.7 21.2 26.3
Made No Significant
Change in Rating 49.6 28.9 33.7 18.5 26.4
47.4 49.6 46.7 38.7
Decreased Rating By
At Least 1 STD. 39.1 56.6 37.6 69.1 54.5 36.5
19.7 32.1 35.0
The table shows the percentage of series to increase the ratings at least one
standard deviation, decrease the ratings at least one standard deviation or to
leave the time slot statistically unchanged. Across the top, the term original
indicates what percentage of the moves resulted in the indicated rating
changed for the original time slot after the established program was moved out.
New indicates the same change for the time slot the established program was
moved into. Series refers to the program itself and shows the percentage of
moved series to change ratings in the indicated direction, while net shows the
same pattern for the networks themselves.
Moving an established series during the viewing year appears to be beneficial
over all to both the series and the network. While it is still true, as can be
seen on Table 1, that the ratings are more likely to significantly decline than
increase (except in the new time slot), This tendency is slightly less than when
the series is left alone. Also the chance for a significant increase in the
ratings is dramatically higher across the board. This suggest that during such
a move the original time slot is able to hold onto its audience while adding
viewers from the replacement series. The moved series itself also was able to
take viewers with it and add them to the audience that was already watching in
the new time slot. As a result, compared to established series not moved,
there is 14.9 percent better chance that the network will significantly increase
its overall ratings, while also improving ratings for the series, the original
and the new time slots.
Moving New Programs
Initial research seemed to provide support for the positions of people such as
Becker, Klein and Barkley. Moving new programs seemed to have very little real
effect as far as improving or harming the ratings was concerned once other
variables were taken into account. While it is true, as shown on Table 2, that
new series that were moved were about 12 percent less likely to be connected
with significant rating declines than other new series, this turned out not to
be related to the move. An analysis of the actual ratings revealed that on the
whole, new series that were moved were already significantly stronger than new
series in general. New series that result in a significant decline in their
original time slot were usually canceled, not moved. As a result, the moved
series tended to have a lower chance of producing a significant decline across
the board. This was not true of established series that were moved, as there
was no rating difference between moved series and those left alone.
Table 2
Comparison of Rating Changes For New Series That Were Moved
Compared To New Series That Were Not Moved
New Series Not New Series Moved New Series Moved
Moved During Summer During Year
Number of Programs 569 series 45 series 190 series
Original New Series Net. Original New Series
Net.
Increased Rating By
At Least 1 STD. 22.5 28.9 17.8 8.9 26.7 26.8
12.6 23.7 26.8
Made No Significant
Change in Rating 34.3 35.6 40.0 15.6 28.9
51.6 56.8 47.4 38.9
Decreased Rating By
At Least 1 STD. 43.2 35.6 42.2 75.6 44.4 21.6
30.5 28.5 34.2
The table shows the percentage of series to increase the ratings at least one
standard deviation, decrease the ratings at least one standard deviation or to
leave the time slot statistically unchanged. Across the top, the term original
indicates what percentage of the moves resulted in the indicated rating
changed for the original time slot after the established program was moved out.
New indicates the same change for the time slot the new program was moved into.
Series refers to the program itself and shows the percentage of moved series to
change ratings in the indicated direction, while net shows the same pattern for
the networks overall rating change.
The real difference between new series that were never moved and those that
were moved can be seen in the "no significant change" row. As the table shows,
this category increased by about 12 percent compared to new series that were not
moved. In other words, the moved series itself, the original time slot and the
new time slot tend to remain statistically unchanged more often than is the case
for new series never moved.
One area on the table does show a large change resulting from the move. There
were 45 limited run series that were renewed and then moved during the period
studied. As the table shows, these series suffered a significant rating loss
75.6 percent of the time. However, further analysis indicated this had less to
do with the move than with the way these series were scheduled to begin with.
All 45 limited run series were used to replaced strong programs for a period of
4 to 6 weeks. During this time, the strong series went on vacation, after which
it returned to the time slot and ran original episodes into the May sweeps
period. The limited run show went off the air. In the fall the renewed
limited run series came back, but in a weaker time slot. The series then
tended to takes on the rating of the new, weaker time slot. The original time
slot tended to return to the rating it was getting before the 4 to 6 week run.
What this indicates more than any thing else is that the new series did not take
on the strength of the strong series, just by being put into its time slot for
a few weeks. As far as the networks were concerned, the practice made virtually
no rating difference.
More in depth analysis of new program moves indicated Table 2 was not the whole
story. New programs do differ enormously in the effects of a move based on how
they were doing in their original time slot. While this variable had no effect
on established programs, it does appear to be a major factor with new series.
Table 3 shows the results when the moved new series were separated by programs
that had significantly improved their original time slot, those that left it
statistically unchanged and series that produced a significant decline in the
original position. As can be seen, 15.4%, or 28 of the new series that were
moved had significantly increased the ratings in the original time slot they
were given. When moved these programs suffered a significant rating decline
53.6% of the time. The network itself suffered a significant rating decline
60.7% of the time. Seventy one new programs, or 39%, had left their original
time slot unchanged in the ratings. Of this group, only 26.8% suffered a
significant rating decline when moved. Finally, 48.2% or 83 new programs had
resulted in a significant decline in the ratings for their original time slot
before being moved. When these series were moved, only 24.1% suffered a further
significant rating decrease. Series in both of these last two categories had a
slightly better chance of a significant rating increase for the program and the
network than a new series that was never moved, while a series that had
significantly strengthened its first slot had less than half the change of an
unmoved series of improving either its own, or the networks rating position.
Table 3
Comparison of Rating Changes For New Series That Were Moved
By The Programs Strength In Its Original Time Slot
New Series New Series That New Series That New Series That
Not Moved Had Significantly Left Original Had Significantly
Increased Slot Slot Unchanged Decreased Slot
Number of Programs 569 series 28 series 71 series 83
series
Series Net Series Net Series Net
Increased Rating By
At Least 1 STD. 22.5 10.7 10.7 25.4
30.6 27.8 25.6
Made No Significant
Change in Rating 34.3 35.7 28.6 47.9 38.9
48.2 42.7
Decreased Rating By
At Least 1 STD. 43.2 53.6 60.7 26.8
30.6 24.1 31.7
The table shows the percentage of series to increase the ratings at least one
standard deviation, decrease the ratings at least one standard deviation or to
leave the time slot statistically unchanged. The headings across the top show
the number of new series that were not moved, the number of new series that had
significantly increased their original slot before being moved, the number to
leave the original slot statistically unchanged and the number to significantly
lower the ratings in their original slot before being moved. The columns show
the percentage of programs under each category to significantly increase,
decrease or to leave the rating statistically unchanged after the move occurred
for both the series and the network.
Multiple Moves and Moves Returning a Series to a Former Time Slot
Moving either a new or established program more than once during any given year
almost always results in a significant rating decline . Eighty four established
series were moved two or mover times during the same year. Sixty nine of those
moves, or 82.1%, resulted in a significant rating decline while only 4.8%
resulted in a significant increase. As far as the networks were concerned, they
suffered a significant rating loss 60.4% of the time, some what less than the
series, but still 21% more often than when the established series was not moved
at all. The chance for the network to improve its overall rating was also just
over 4%, or substantially below the control condition. It made no difference
whether the first move had taken place in the summer or during the main viewing
year. The second moved resulted in the same extreme decrease.
During the period studied, 39 new series were also moved more than once during
the year. Of these programs, 26 were moved twice and resulted in a significant
rating decline 71.8 percent of the time with a 5.1% chance of a significant
increase. The network also suffered a significant decline just over 63% of the
time. The remaining 13 programs were moved three of more times and declined
significantly in the ratings 100% of the time. The network also declined
significantly 100% of the time. Researchers reasoned that multiple moves might
have occurred as a result of program weakness or from the failure of the first
move. However, analysis of the established series found no difference in
ratings, gains or loses between programs moved once or more than once. New
series moved more than once were more likely (just over 12%) to have improved
the ratings for the series and the network after the first move. This might
suggest having succeed with one move, programmers tried to do it again, usually
with very bad results.
Between 1986 and 1997 moves resulted in 57 programs being returned to their
former time slot. When this happened, 29.8 % of the series returned to their
former rating, and 38.6% of the time the network improved its overall rating
significantly. However, 28.1% of the time both the network and the series
declined significantly. In other words, there is some ability for a program to
regain its former rating status if it is put back into its original time slot,
but there is an almost equal chance the return to the original slot will result
in further significant rating losses.
Discussion
The study indicates all program moves are not created equal. It also indicates
the process of moving series does effect the ratings. Moving an established
series once during the viewing year actually seems to improve the chance for a
significant rating increase for both the series and the network, however, moving
an established series during the summer usually hurts both the network and the
program significantly. This rises real concerns about the common practice of
shifting established programs around when the season starts in the fall in order
to make stronger nights, lead-ins or blocks. This data provides strong
indication that such strategies don't work.
How a move effects a new series seems to depend on how the new series was doing
in the first place. For a new series that harmed its original slot - i.e. lost
rating points, or a new series that made no change in the audience, a move
seems to give the show a second chance. However, if the new series had
significantly improved the slot it was first given, a move was pretty much the
same as a death sentence.
The ratings for either new or established series don't survive multiple moves
during the same year and, while the effect of moves on networks and programs are
not the same, such multiple moves also do not benefit the network. This would
seem to suggest this common practice should be avoided., regardless of what
strategy is being created.
While there is some limited ability for a program to reestablish itself when it
is returned to its former time slot, that ability is "limited." In short, it is
not that easy to repair rating damage once it has occurred. Programmers cannot
count on correct unintended loses just by undoing what they did in the first
place. These findings provide strong backing for qualitative results indicating
viewers become frustrated by too many moves and leave.
The study indicated schedule instability is more than just a method to reach
other programming goals. The moving itself does effect the numbers. This would
suggest programmers need to think carefully about which shows they move and why
they are doing it. Such moves can indeed be beneficial to both the series and
the network, but only under specific conditions. Moving just to tweak the
schedule would seem to be a dangerous idea. Indeed, this study provides enough
indication that scheduling instability may be a factor in the loss of audience
to warrant further research into the practice of moving programs and to question
the advisability of quick cancellations or the heavy use of specials which
result in a breakdown of normal programming patterns. Programmers cannot assume
such practices are beneficial, or, at the very least, not harmful. This constant
instability may very well be a factor in why the broadcast networks are losing
the audience.
Adams, W. J. (1999). How people watch television as investigated using focus
group techniques, Journal of Broadcasting & Electronic Media, forthcoming.
Adams, W. J. (1997). Scheduling practices based on audience flow: What are the
effects on new program success? Journalism and Mass communication Quarterly,
74(4), 839-856.
Adams, W. J. (1993). Program scheduling strategies and their relationship to new
program renewal rates and rating changes, Journal of Broadcasting & Electronic
Media, 37(4), 465-474.
Adams, W. J. and Eastman, S. (1997). Prime-time network television programming.
In S. Eastman & D. Ferguson, (Eds.), Broadcast/Cable Programming Strategies
and Practices. 5th Edition (122). Belmont, CA: Wadsworth Publishing Company.
Boemer, M. L. (1987). Correlating lead-in ratings with local television news
ratings, Journal of Broadcasting & Electronic Media, 31(1), 89-94.
Brosius, H., Wober, M. & Weimann, G. (1992). The loyalty of television viewing:
How consistent is tv viewing behavior? Journal of Broadcasting & Electronic
Media, 36(3), 321-335.
Cooper, R. (1996). The status and future of audience duplication research; An
assessment of rating-based theories of audience behavior. Journal of
broadcasting & Electronic Media, 40(1), 96- 111.
Gitlin, T. (1983). Inside Prime Time. New York, NY: Pantheon Books, pp. 60-62.
Klein, P. (January 25, 1971). The men who run tv aren't stupid. . .They know use
better than you think. New Yorker Magazine, pp. 20-29.
Lin, C. A. (1995). Network prime-time programming strategies in the 1980s.
Journal of Broadcasting & Electronic Media, 39(4), 482-495.
Livingstone, S. (1993). The rise and fall of audience research: An old story
with a new ending. Journal of Communication, 43, 5-12.
Morley, D. (1993). Active audience theory: Pendulums and pitfalls. Journal of
communication, 43,
13-19.
Tiedge, J. T. & Ksobiech, K. J. (1987). Counterprogramming primetime network
television. Journal of Broadcasting & Electronic Media, 31(1), 41-55.
Wakshlag, J. & Greenberg, B. S. (1979). Programming strategies and the
popularity of television programs for children. Human Communication Research,
6(1), 58-68.
[1] Information on program moves were taken from Broadcasting & Cable magazine
checked against scheduling information in Variety for 1997 and 1998.
|
