Time = Money

1
Time Money

• Freya Caudwell, Beki Hardcastle, Sarah Fraser
  Alan Pye

2
When time is money the effect of hourly
payment on the evaluation of time

• Composed of three related studies.
• All studies relate to whether or not people paid
  hourly develop different conceptions of time in
  comparison to people paid a salary, with specific
  regards to how much they view time as being worth.

3
Study 1

• Built on a study by Soman(2001).
• Analysed whether people paid hourly wages held
  stronger to the view of time in monetary terms,
  compared to those who were not paid hourly.
• Predicted people paid an hourly wage would be
  more likely to agree to statements of mental
  accounting relating to viewing time in monetary
  terms.

4
Study 1 method

• 88 participants.
• Anonymous questionnaire. 86 of participation
  rate.
• 10 ppts excluded from analysis.
• 11 ppts paid hourly, rest salaried.
• Rated endorsments of 8 statements of mental
  accounting from Soman(2001) on a 1-9 likerk scale.

5
Study 1 results

• Calculating hourly wage had an effect on how ppts
  viewed money(p .02)
• Hourly participants did not vary from salaried
  participants in response to statements (p.4)
• However, they were more likely to endorse mental
  accounting model for time (p.07)
• Hourly ppts scored higher on every individual
  question on Somans scale (p.008)
• No difference directionally between salaried and
  non salaried with respect to monetary response.

6
Paper 1, study 1 - critique

• Sample possibly non-representative.
• Utilised opportunity sampling.
• 11vs54 ppt subgroups, possible problems relating
  to statistical power.
• Time constraints.
• Waiting for train might well make economic
  factors more salient.
• Work by the hour jobs might attract a specific
  type of person.
• Didnt ask about previous experience of working
  jobs that pay by the hour.

7
Study 2

• Building on the previous study attempting to
  determine what other factors might mediate
  viewing time in a monetary sense and determine
  the effect sizes of the relevant effects.
• Used publicly available survey data
• Controlled for factors that might affect the
  experiment earnings, marital status, education
  etc.

8
Study 2 method

• May 2001 current population survey work schedule
  supplement was used (CPS 2001).
• Most recent datasheet to include items assessing
  willingness to trade time for money
• Ppts were asked if they would work less time for
  less money, the same hours for the same amount,
  or more hours for more money.
• Weekly earnings, amount of hours worked, and
  sector of job were controlled for.

9
Study 2 results

• Most salient variables for willingness to trade
  time for money were current earnings and hours
  currently worked.
• Hourly workers earned less, and worked less long.
• Significant differences between hourly and
  salaried (plt.0001). 32.1 of hourly vs 17.1 of
  salaried were willing to trade.
• Hourly payment associated significantly with
  willingness to trade (p.003) according to a
  regression equation, including control variables.
• Effect size of hourly payment was comparable to
  actual number of hours spent working (a much more
  significant effect (plt.0001)).

10
Paper 1, study 2 - critique

• Huge sample size of 13,000
• Only measured current payment conditions, not
  life experience of being paid by the hour.
• Many variables controlled for.
• Disinterest in survey might affect results.
• Variation in who responds to survey.
• No variable manipulation cant state causation.

11
Study 3

• Analysed the effects of lifetime exposure to
  hourly wage on viewing time in a monetary sense.
• Determine whether or not making salaried people
  view their wages in terms of payment per hour
  would lead to them viewing time in a more
  monetary sense.
• Attempting to factor in the variable of how much
  economic criterion relate to their day to day
  decision making processes.

12
Study 3 - method

• 62 Ppts (39 female) from nationwide database at a
  university.
• Online questionnaire paying ppts 5. Ages ranged
  from 19 to 67.
• Ppts responded to three questions about how much
  they earned before taxes, how many hours they
  worked per week, and how many weeks they worked
  per year, in the previous year.
• Some participants were asked to calculate their
  hourly wage.
• A 7 point Likert determined how much experience
  ppts had with working for hourly wages, and how
  much more or less time they would spend if given
  the option on paid work.
• Ppts completed three items determining the extent
  to which they considered economic criteria
  salient when making decisions, determined with a
  7 point Likert scale.

13
Study 3 results

• Prior exposure to hourly work was associated with
  willingness to spend more time on paid work
  (p.01).
• A main effect of calculating an hourly wage was
  not found, but there was a significant
  interaction between the experimental condition
  and prior hourly experience (p.04).
• Prior exposure to hourly payment was associated
  with a greater salience of economic factors
  (p.002).
• A mediation analysis revealed that the
  interaction between hourly payment experience and
  willingness to spend more time to make more money
  was mediated by the salience of economic factors
  in decision making.

14
Paper 1, study 3 - critique

• Ppts numbers were low, although they were
  selected from a large group.
• Participation reward was given as a voucher for a
  specific retailer.
• No main effect of calculating hourly wage.
• Mediation, but yet again insufficient data on
  other variables that ought to be included.
• No true controls.

15
Time, money, and happiness. How does putting a
price on time affect our ability to smell the
roses?

• Composed of three related studies
• All studies relate to how the impatience that
  stems from putting a price on time impairs
  ability to derive happiness from enjoyable
  experiences.

16
Study 1

• Analysed the effect of an unexpected period of
  leisure time on the computer on the ppts
  happiness.
• Predicted ppts prompted to consider the monetary
  value of their lost time would derive less
  happiness from it than unprompted participants.

17
Study 1 - method

• 53 undergraduates (32 female) as ppts, all
  received 5.
• Upon arrival, completed questionnaires, and then
  were given 10 minutes leisure time on the
  internet, and completed a happiness rating at the
  end.
• Answered questions on their employment
  expectations post university.
• Some participants calculated their expected
  hourly wage, others did not.
• Assessed ppts happiness with a 5 point Likert
  scale question, prior to experimental
  manipulation and post leisure period.

18
Study 1 results

• 22 repeated measures ANOVA.
• Participants asked to calculate their hourly wage
  were no happier after the leisure period than
  before (p.43), but ppts who did not were happier
  post leisure period (p.01).

19
Paper 2, study 1 - critique

• 5 participation payment.
• Only 53 ppts, that might not be generalizable.
• Given ppts were in further education talking
  about future wages, they were likely to be overly
  ambitious.
• Ppts were asked Do you feel happy in general
  - this might not capture the variations between
  peoples emotions (Clark et al, 1988, Lepper et
  al, 1999, Zhang et al, 2002).
• Being prompted to think about time might have
  affected browsing habits.
• No mention of R2 or R in regression, or altered
  R2. This limits the information we can gain from
  the regression equations.
• No pure control condition.

20
Study 2

• Hypothesised that thinking about the monetary
  value of time induces a sense of impatience with
  the leisure period, and that this is the
  mediating effect on happiness.
• Utilised listening to music instead of leisure
  period to induce happiness.

21
Study 2 - methods

• 401 ppts, 237 female, recruited from Amazons
  mechanical turk worker pool, in exchange for 1.
• 39 of these ended up being excluded.
• Used the same questions about income as before,
  but this time it referred to the income in the
  previous year, instead of expected income.
• Added a pure control condition, with no mention
  of work or income.
• Responded to a similar question on happiness, but
  this time with a Visual Analogue Scale (VAS)
  (Horne et al, 1988).
• Six self report items used to measure impatience.

22
Study 2 results

• Ppts in the experimental condition experienced
  less happiness after the music than either the
  control or the pure control group (p.03)
• Ppts in the experimental condition experienced
  more impatience listening to the music than the
  control or the pure control group (p.003).
• Mediation analysis revealed that the effect of
  thinking about hourly wage on happiness was
  mediated entirely by impatience.

23
Paper 2, study 2 - critique

• Impatience is a poorly defined variable it
  could be misreported for due to social
  desirability and demand characteristics.
• Impatience items had high Cronbachs alpha.
• Using VAS for the impatience items artificially
  conflates Cronbachs, since people seem unlikely
  to have too much variation.
• Uses mechanical turk, a unique subject pool that
  is unlikely to be overly generalizable(Ross et
  al, 2010, Gosling et al 2011)
• Utilises a lot of T-tests, with no mention of
  checking assumptions.
• In one T-test, determined no difference between
  groups based on T(370) 1.56, p.119. This seems
  overly ambitious, given it affected later
  analysis.

24
Paper 2, study 2 - critique

• Used music instead of internet.
• Measured happiness with a different metric to
  that of study 1. The question was different Do
  you feel happy in general compared to How happy
  do you feel right now. They were also measured
  on different scales.
• Talks about impatience mediating the effect of
  calculating an hourly wage on happiness.
• No happiness test before music, therefore could
  be measuring from a baseline of pre-music
  happiness which might differ between groups.

25
Study 3

• Wanted to extend study 2.
• Hypothesised that being paid for listening to the
  music would have a direct effect on the happiness
  of those people who are considering their hourly
  wage.
• Would have no effect on the happiness of those
  people who have not considered their hourly wage.

26
Study 3 - method

• 205 ppts, 117 female recruited from Amazons
  mechanical turk in exchange for 1.
• 21 of these were excluded from final analysis,
  and a further 39 were excluded because they were
  part of study 2.
• Control conditions were replaced with making ppts
  carry out meaningless arithmetic equations.
• Half of ppts were explicitly compensated 50 cents
  for the time spent listening to the music.
• Two VAS scales were introduced to measure ppts
  enjoyment of the music.

27
Study 3- results

• Ppts paid to listen to the music enjoyed it more
  than those who were not (p.04).
• Ppts who calculated hourly wage and were
  compensated derived more enjoyment from the music
  than those who were not compensated (p.004).
  Control ppts did not differ significantly
  deepening on compensation however, (Flt1).
• In no compensation groups, enjoyment was
  significantly lower in the group required to
  consider their hourly wage (p.09).
• A mediation analysis indicated that the hourly
  wage vs happiness relationship became
  insignificant when impatience was considered.
• Impatience also predicted happiness significantly
  (Plt.001).

28
Paper 2, study 3 - critique

• Utilised mechanical turk, which might be a poor
  worker pool.
• Excluded ppts who participated in study 2,
  preventing possible confounds from that.
• Made repeated subdivisions of groups into
  conditions. This might have sacrificed
  statistical power.
• Implied significance of an effect at p.09.
  Arguably, this is not significant, since p
  /lt.05, however, due to the reasons stated
  above, this is likely due to sample subdivisions,
  and not a lack of an effect (Goodman, 1999).
• Had questions to test for enjoyment of music,
  which possibly accounts for some critique
  relating to musical taste.

29
Any Questions?