The 1997 Storting election clearly demonstrates that Norwegian politics remains in a turbulent and volatile state. It was perhaps even more special than previous elections, and in many ways became a record-breaking election (Aardal 1998). To some extent, all elections have their distinguishing marks. The 1989 Storting election, for instance, marked the end of an eventful decade that began with the swing to the right in the early 1980s (Valen, Aardal & Vogt 1990, 110). The 1993 Storting election showed a completely different tendency, distinguished by an intense debate about our relationship with the European Union (Aardal & Valen 1995, 10, 227-28). In the early 19905, Norwegian politics entered a state of change wherein old ties were broken and new patterns had not yet been set (ibid., 227-35). However, this tendency was not entirely new. Even in the early 19705, traditional ties

Resumé

Norwegian politics remains in a turbulent and volatile state. The 1997 Stoning election became a record-breaking election where two parties, the Christian Peoples Party and the Progress Party, achieved their best ever results. The Conservative Party experienced the worst result in the party's more than hundred years' history. The Labor Party had its second worst election since the 19305, while the Center party halved its number of voters and lost two thirds of their MPs compared to the previous election. The election campaign played a decisive role in this outcome. More than half of the voters decided which party to vote for during the campaign. Lack of commitment, rather than the parties' ability to create a positive interest in the election, seems to have caused the large shifts in support for the parties and the record-high share of voters who decided during the campaign. 43 percent of the voters changed party preference from 1993 to 1997. If we exclude non-voters, 33 percent switched party. The main issues of the campaign were health and eldercare. Although these issues dominated in all party groups, we find clear evidence of "issue ownership," where specific parties attract voters with particular agendas and issue priorities. The Labor Party government stepped down after the election and was replaced by a centrist government led by Prime Minister Kjell Magne Bondevik. The new government is one of the weakest minority governments in Norwegian history and is only supported by 42 of the 165 Storting members.

---

* Bernt Aardal, The Norwegian Election Studies, Institute for Social Research, Munthesgt. 31, N-0260 Oslo, Norway. E-mail: bernt.aardal@isaf.no

between voters and political parties were dissolving, and more and more
voters shifted party preference between elections (Valen 1981).
In the 1997 election, two of the parties, the Christian People's Party and the

Progress Party, achieved their best ever results. The Conservative Party experienced the worst result in the party's more than hundred years' history. The Labor Party had its second worst election since the 19305, while the Center Party halved their number of voters and lost two thirds of their MPs compared with the previous election. The Red Electoral Alliance experienced a better result than in 1993, but still lost their only MP. In Nordland county, the outsider Steinar Bastesen succeeded in his bid for parliament as a representative for Kystpartiet (Coastal Party).

If the election results were remarkable, so was the election campaign. Before the campaign started, it seemed certain that the Labor Party with its new Prime Minister, Torbjørn Jagland, would stay in office after the election. The opposition was still divided after the conflict regarding membership in the European Union and numerous economic factors seemed to benefit the ruling government as well. Inflation and interest rates were low, unemployment had been drastically reduced and was among the lowest in the western hemisphere and the budget surplus was increasing. Norway was as an economic oasis compared to most industrial countries, largely due to the enormous revenue from the oil industry. The oil revenue had previously been used on investments and to pay debts, but now it was placed in oil funds that increased with staggering figures. When Gro Harlem Brundtland left her post as prime minister to Torbjørn Jagland in October 1996 - barely a year prior to the election - she had prepared the ground for success at the polls. Jagland's governing period, however, turned out to be more problematic than expected. First, Minister of Planning and Coordination Terje Rød-Larsen had to resign after only a few weeks, due to a ten-year-old tax case. Then, Minister of Justice Grete Faremo was compelled to resign due to problems concerning investigations of the Secret Service. Finally, Faremo's replacement, the untraditional Anne Holt, had to withdraw due to poor health. In addition to problems with his staff, Jagland's attempt to set a new agenda by introducing the concept of "the Norwegian house" was ridiculed by the opposition and the media. The opposition seized on every opportunity to impair the new Prime Minister's authority, for instance when they enforced unprecedented increases in the state budget. In addition to the budget increase, the opposition, except the Conservative Party, proposed a considerable increase in the minimum old age benefits. This proposal was strongly criticized by the Prime Minister who claimed that the opposition's outbidding of the government was an irresponsible flirt with the voters. This happened in June 1997 at the end of Stortinget's spring session. What nobody knew at the time was that this had prepared the ground for one of the major issues of the upcoming campaign, namely eldercare.

Election Results

Table 1 reports each party's share of the votes in the 1977 to 1997 elections, while Table 2 shows each party's number of seats in the same period. A special point of interest before the last election was whether Norwegian politics would return to its "normal" state after the bitter conflict about the European Union, as it did after the EEC referendum in 1972 (Valen 1976; Christophersen 1976). The Center Party's marked decline from 16.7 percent of the votes in 1993 to 7.9 percent in 1997 could indicate that Norwegian politics once again had returned to a "normal" state, as this party was the most prominent anti-EU advocate. Even if the Center Party's result in 1997 was better than the party's results before the EU entered the agenda, a lot of the support it received in connection with the EU debate had dissipated. A closer look at the results for each municipality confirms that there was a strong correlation between the Center Party's decline in 1997 and anti-EU support in 1994. The Center Party lost more support in municipalities with a large number of no voters in the referendum than they did in other municipalities.

An important part of the "normalization" in the 1970s was the success of the Labor Party and the Conservative Party in winning back voters they had lost due to the EEC conflict. The two parties received in excess of 66 percent of all votes in the last election before the referendum in 1972. In the first election after the referendum, the same parties received less than 53 percent of the votes. Four years later, however, the two parties were back to where they were before the referendum. In 1977, they received 67 percent of the votes and in 1981 as much as 71 percent of the votes. This situation had changed before the last referendum. The two parties received just under 57 percent of the votes in the 1989 Storting election. Their share of the votes dropped to 54 percent in the 1993 election, which was dominated by the EU conflict. An important reason why the swing was less dramatic than in the 1970s was that the election this time came one year before the referendum while in the previous case, the election came one year after the referendum. In the 1997 election, however, the Labor Party and the Conservative Party were supported by only 49 percent of the voters. This was the first time since World War II that these two parties together received less than half of the votes cast in a national election. Hence, insofar as the level of support for Labor and the Conservatives is part of a "normal" pattern, Norwegian politics has not returned to "normalcy." Nor can the results of the Progress Party and the Christian People's Party be said to fit into a "normal" pattern.

Tables 1 and 2 clearly indicate significant fluctuations between elections
for all parties over the last two decades.

Three Phases of the Campaign

The 1997 election campaign can be split into three different phases. The first phase began with the closing of Stortinget's spring session. The proposal to raise the minimum old age pension received so much attention that the unsolved problems of the welfare state completely overshadowed the positive economic trends, which the government would have preferred to focus on. The portrayal of Norway as a wealthy oil nation may even have intensified dissatisfaction with shortcomings in the welfare sector. The contrast between public and private wealth on one side and growing waiting lines for operations on the other may have made the latter problem even less acceptable for many voters. An opinion poll taken in late April and early May by AC Nielsen showed that health issues and eldercare were the single most important issues among the voters. The same poll showed that the Progress Party did especially well among voters who had these issues as their top priorities. A poll from Norsk Gallup in late June, which gave the Progress Party 22 percent of the votes, sent shock waves through the political establishment. Subsequent polls confirmed that the Progress Party had indeed

won a lot of support, and some polls gave the party as much as 25 percent of the votes by mid-August. This caused panic among the other parties, and Torbjørn Jagland and the Labor Party immediately changed their campaign strategy and now named the Progress Party as their main opponent. Earlier that spring, Labor and the Conservatives had appointed each other as main opponents in concordance with the traditional portrayal of "enemies" in Norwegian politics. Now, Jagland claimed that the Progress Party represented the "new Conservatives" and sought a confrontation with the party in a live TV broadcast. The Labor Party's attack was primarily linked to the Progress Party's attempt to present itself as the defender of underprivileged groups, such as the sick and the elderly. The party's attitude towards immigration policy received less attention than their policy on welfare issues. Jagland's challenge served two purposes: First, it put the Labor Party on the offensive after having had to defend itself so far in the campaign, and second, it put pressure on the Progress Party. Both during and after the debate, it seemed like the Labor Party succeeded in getting the Progress Party on the defensive. The polls showed progress for the Labor Party in August, while the Progress Party suffered a marked decline. The election showed that the Progress Party received 10 percentage points less than the 25 percent the polls had given them earlier in the campaign.

The campaign's second phase began in September when Prime Minister Jagland's position was challenged by Kjell Magne Bondevik. The three centrist parties - the Christian Peoples Party, the Center Party and the Liberal Party - had already agreed to present themselves as an independent alternative to the Labor government. The new thing about this centrist alternative was that the Conservative Party was not included as part of a non-socialist alternative. This was due to the EU conflict and the ensuing political gap between the Center Party and the Conservative Party. The two other centrist parties were more willing to cooperate with the Conservative Party and did not dismiss that the Conservative Party might, at some point, become part of a coalition government. The three parties had, however, agreed before the election to pursue the centrist parties as an independent alternative.

In the further course of the campaign, Bondevik's personal popularity became the focus of attention. Shortly before the election some polls even indicated that Bondevik was a more popular candidate for prime minister than Torbjørn Jagland. In consideration of their respective parties' size and the voters' apparent skepticism towards the centrist alternative, Bondevik's popularity was remarkable. Parallel to Bondevik's popularity, there was also increased support for the Christian People's Party. This development could particularly be seen in the final stages of the campaign.

How the two other centrist parties would perform seemed less certain. The
Center Party, who had kept its support from the local elections in 1995 until
the spring of 1997, suffered a marked decline in the closing stages. The

Liberal Party also had problems in the fall. Evidently, the popularity of the
Christian People's Party and Mr. Bondevik did not increase overall support
for the centrist alternative.

The campaign's third phase covers the last few days before the election. Torbjørn Jagland had brought excitement into the final stages of the campaign with his "question of confidence to the voters." He threatened that if Labor failed to receive at least the same level of support as in the last Storting election (i.e., 36.9 percent), the government would resign. During the campaign, Labor had managed to bounce back from a declining trend, and the polls indicated that Jagland was close to reaching his goal. The party leader debate on Friday before the election may have contributed to prevent this. Usually, the party leader debates are considered a national ritual with minute influence on the results (Semetko 1996, 227). However, Prime Minister Jagland's effort in the debate, in which he, among other things, was accused of lying, caused further uncertainty about the election results. Several polls showed a decline in support for the Labor Party in the last few days of the campaign. Thus, the hope of achieving 36.9 percent of the votes collapsed in the final lap. Labor missed the goal by only 1.9 percentage points.

The 1997 campaign was in many ways a more dynamic and unpredictable process than the previous campaigns. The 1997 Election Study confirms this impression.¹ Figure 1 shows the percentage who answered that they decided which party to vote for during the campaign or just prior to election day. The

question was formulated as follows: "When did you decide which party you were going to vote for? Was it long before the election campaign began, was it some time during the campaign, or was it just prior to election day?" Figure 1 shows the share that chose one of the last two alternatives.

The figure shows that the share of voters who made up their minds during the campaign had a distinct increase in the late 1980s. In 1989 and 1993, more than 40 percent of the voters were influenced by the campaign when they decided which party to vote for. In the last election, this share had risen to over 50 percent! Among those who decided during the campaign, 29 percent answered that they made up their minds just prior to election day. The overall increase mainly occurred among those who made up their minds earlier in the campaign. If we compare these figures with the polls taken during the campaign, it becomes clear that the campaign was decisive for the outcome of the election. It is noteworthy that the share of voters who make up their minds during the campaign seems to increase in a step-like pattern: From its lowest level in the 1960s to a higher level in the 19705, where it reached a plateau, only to increase in the late 80s, and again in the last election.

Voters were also asked how significant they thought the campaign on TV, radio and in the newspapers was for how they voted. 16 percent answered that the media's campaign coverage was very important or quite important, 37 percent answered that it was somewhat important, while 47 percent answered that it was not important. In other words, more than half of the voters answered that they were influenced by the campaign. There is also a clear tendency that the more important the campaign, the longer people waited before deciding who to vote for.²

The question is whether the fact that the voters were influenced by the media's coverage of the campaign also means that they became committed to, perhaps even enthusiastic about, the campaign? Table 3 shows that the answer is no - commitment was definitely low.

Only 35 percent of the voters reported that it was very easy to decide which party to vote for, while 28 percent "cared a great deal" about the election outcome. Furthermore, only 24 percent said they discussed the election daily.³ Compared with the four previous Storting elections, commitment is generally lower for all three questions. All in all, the 1997 election was

definitely not a mobilizing election. Lack of commitment, rather than the parties' ability to create a positive interest in the election, seems to have caused the large shifts in support for the parties and the record-high share of voters who decided during the campaign. If we look closer at the election's winners, the Christian People's Party and the Progress Party, we find that although their voters had an easier time making up their minds, the difference between these voters and those of the Labor Party or the Center Party in terms of decisiveness is negligible (table not shown here). The voters of the Christian Peoples Party cared slightly more than average about the election result, but so did those of the Socialist Left Party, Labor and the Conservative Party - parties that in no way can be called election winners. As to the question of how often people discussed the election, we find that the Progress Party's voters are somewhat above average (29 percent), while the Christian People's Party's voters are clearly below average (19 percent). When the parties' mobilization is divided into three periods - long before the campaign started, during the campaign and just prior to the election - the parties' share of the three voter groups is as follows:

The Socialist Left Party got more support from voters who decided just prior to the election than in the other two periods. The Labor Party got less support from voters who decided just prior to the election than those who decided during the campaign, and even less compared to those who had made up their minds long before the campaign started. The Liberal Party got more support just prior to the election than in the other periods. The Christian People's Party performed relatively best during the campaign, but also won support just prior to the election. The Center Party got most support from those who decided before the campaign had started, less from those who decided just prior to the election, and least among those who decided during the campaign. The Conservative Party had the best period just prior to the election, while the Progress Party had their worst period just prior to the election.

What Was the Campaign Really About?

Single issues tend to mobilize voters to the polls and to have a significant impact on their decision who to vote for, which has created a greater interest in the mass media's performance in this process. In an age when the parties' grip on the voter has become weak, the media occupies an important intermediary position as "interpreters" of the political processes. In this connection, the media's agenda function has been stressed (McCombs & Shaw 1972; Iyengar & Kinder 1987), but there are also theories that claim that the media has a direct impact on the voters' attitude formation. Noelle theory of the "spiral of silence" emphasizes that individuals who

find support for their own attitude in the media attain more self-confidence and are more active in the public debate, while individuals who belong to a minority in terms of attitudes and issues withdraw from the public debate and become ever more passive. The first case produces an upward-confirming spiral, and the latter a downward spiral - a spiral of silence (Noelle-Neumann 1974; 1977; 1984). We cannot go into detail about the media's significance for attitude formation and issue interest here, but the agenda setting function and the theory of the spiral of silence emphasize that the mass media plays a decisive role - especially during election campaigns. Over time, the media's role has changed from an arena for parties and politicians into an independent agent (Esaiasson 1991, 261-78; Bjørklund 1991, 279-302). Given the media's dominating role, it would seem reasonable to expect that the voters do not have a clear perception of which political issues are most important before the media has fully begun to cover the election campaign. This implies that the media sets the agenda and the voters tag along. It would also be reasonable to expect that the voters' interest in different issues will concentrate and narrow considerably during the campaign. In line with the spiral of silence theory, there will be less room for minority issues as the campaign progresses. Although we have no data that allow us to study the direct relationship between the media's and the voter's agenda, it can still be interesting to use these assumptions as a starting point for the analysis of which issues distinguished the 1997 election campaign.

We have documented that the election campaign failed to create enthusiasm among the voters. With this in mind, one might suspect that only a few, if any, issues were of interest to the voters. But, as already mentioned, health and eldercare were in focus at an early stage of the campaign. Both the Labor Party and the Progress Party had these issues at the top of their campaign posters. Statistics Norway's May-June omnibus confirms the pattern from the AC Nielsen poll. Eldercare topped the voter's list, with health issues close behind, and the voters' concern with these issues increased slightly during the campaign.

It is nothing new that Norwegian voters stress health and social issues. Already in the local elections in 1983, these issues were listed among the most important ones (Valen 1984, 27), and polls indicated that they were the two most important issues in the 1985 and 1989 Storting elections as well (Aardal & Valen 1989, 44; Valen, Aardal & Vogt 1990, 16). In 1993, eldercare and health issues were ranked as number three, right behind EU membership and unemployment. After the question of membership was decided in the 1994 referendum, and unemployment had been drastically reduced, there was room for other issues. Considering the priority the voters had given to health and the elderly in previous elections, it was not unexpected that precisely these issues topped the agenda.

Although the campaign turned out to be less about environmental issues than expected, a relatively large number of voters answered that they were concerned about the environment as well as education, child and family issues, which became more prominent in August and September than in May Once again, the voters' priorities reflect what was going on in the campaign. Interest in the immigration issue was halved from May until September. Evidently, the Progress Party's success at the Storting election had other causes than the immigration issue, which was far more prominent at the local elections two years earlier (ef. Aardal, Valen & Berglund 1996). It is worth noticing that the Conservative Party's traditional issues, e.g., tax policy, ranked poorly with the voters, and the Center Party's main issues, EU and district policy, rank poorly already in May-June and stayed low throughout the campaign.

A question about voter priorities was also asked in the post-election survey. Table 4 reports the priorities of the issues just after the election compared with corresponding data from the 1993 Election Study. The table confirms that issues considered important early in the campaign held their status throughout the campaign.

Health issues and eldercare were clearly at the top of the voters' priority
list. Compared to the previous Storting Election, concern with these issues

more than tripled. A relatively large number of voters also mentioned child and family issues and education as important, and environmental issues and social equalization are also mentioned quite often. Notice that the share of voters who mentioned the main issue in the 1993 election, the EU, dropped from 65 percent to 4 percent. Likewise, the share mentioning unemployment drops from 22 percent to 7 percent. As we have already mentioned, the election campaign was decisive for the final result. It is interesting, then, that the main issues of the campaign were present in the voters' minds well ahead of the final stages of the campaign. The changes in the priority of issues during the campaign consist of an enhancement of some issues and a weakening of others. This suggests that the media - which plays a vital role in the final stages of the campaign - does not necessarily set its own agenda, but contributes to strengthening or weakening existing tendencies. At the same time, the total number of issues mentioned is relatively high, not only before the campaign started, but also close to election day. Despite the dominance of health issues and eldercare, a variety of other issues are mentioned.

Issue Ownership

Although Table 4 clearly shows which issues were most important for the campaign, the number of issues is so large that there is room for considerable variation in the priority of issues among various groups. Ansolabehere & Iyengar (1994) have introduced the concept of "issue ownership," meaning that some parties seem to "own" certain issues in the sense that the voters have an understanding of which parties are best at dealing with these issues (ibid.; Petrocik 1996). In Norway, an example of this is the voters' faith in the Labor Party's ability to combat unemployment and the faith in the nonsocialist parties' ability to combat inflation (Martinussen 1972, 154-55; Valen 1981, 320-24). A number of studies have shown that different voter groups often have quite different priorities and that the priority of issues within each group is quite stable over a period of time (Valen & Aardal 1983, 45; Aardal & Valen 1989, 48^9; Valen, Aardal & Vogt 1990, 17; Aardal & Valen 1995, 21). The tendency that parties seem to own certain issues is partly due to their place in a historically set conflict structure, partly to their focus on issues over a period of time, and finally to the voters' experience of how the parties have handled the issues in everyday politics. It is important to note that the tendency towards issue ownership is especially valid during election campaigns (Petrocik 1996, 826). In many ways, this is a question of whether we are dealing with a national and uniform political agenda, or if we have a number of different agendas and issues. Table 5 reports the share of voters that mentions the different issues by party preference.

Table 5 shows that health and eldercare had top priority in almost all parties. The Progress Party's voters distinguish themselves by prioritizing these issues more than any other party's voters. We notice, however, that also voters of the Labor Party, the Christian People's Party and the Conservative Party often mention health issues and eldercare as their top priorities. As far as these issues are concerned, we can conclude that we are dealing with a national, common agenda. For the remaining issues the picture is less clear. We find that child and family issues are especially important to the Christian People's Party's voters, which was to be expected since the proposal of kontantstøtte (cash support for the care of small children) put its mark on the campaign. The Socialist Left Party's attempt to profile itself on education also seems to have been successful, although the table says nothing about how many new voters this got the party. We also notice that environmental issues appeal to voters from the Socialist Left Party and the Liberal Party. Social equalization is prominent within the Socialist Left Party. When we look at economic issues, we find that the voters for the Conservative Party and the Progress Party are most prominent, closely followed by the voters for the Labor Party, who are also the only ones to show much interest in unemployment. Although immigration ranks low on the agenda, we find that the voters of the Progress Party were truly concerned about this issue. The

same is true for the Center Party's voters and their concern about district policy and the EU. The Liberal Party's focus on industrial and tråde policies is reflected by its voters. Finally, moral-religious issues are of special interest almost exclusively to the voter's for the Christian People's Party. Table 5 demonstrates that in addition to a national agenda there are a number of "local" agendas. In connection with previous studies this indicates that the parties have in faet managed to maintain their issue ownership - even when "their" issues are not on top of the campaign agenda. While only 9 percent of all voters had tax issues as their top priority, for instance, 19 percent of the Conservative Party's voters and 15 percent of the Progress Party's voters had this on top of their list.

Concluding Remarks

Although the last Storting elections all seem to have been "different" and "special," they have at the same time signified long-term processes of social and political change. The high level of volatility from one election to the next represents a trend that can be observed in numerous other countries (Crewe & Denver 1985). In Norway, panel data show that 43 percent of the voters changed party preference from 1993 to 1997. This is on par with the recordhigh volatility of the 1993 election (44 percent). If we include only voters who had voted at both elections, i.e., excluding non-voters, we find that 33 percent switched party, which is also about the same as in 1993. Obviously, the Norwegian party system is still living in the shadow of the EU controversy, although the main contenders of that debate have lost out. On the other hand, Prime Minister Jagland's threat materialized. His government resigned just after the election and was replaced by a minority government consisting of the three centrist parties, led by Prime Minister Kjell Magne Bondevik. However, the new government controlled only 42 of the 165 members of Stortinget, one of the weakest minority governments in Norwegian history. At this writing, the new government has been in office for almost eight months. No matter how long the government will be in office, it represents a new alternative in as much as it tries to maneuver independently of the traditional "socialist" and "bourgeois" blocks. Even if the centrist government is toppled during the present election period, the strength of the various parties will remain the same, as there is no provision in the Constitution for dissolving Stortinget and calling for new elections. Consequently, Norwegian politics will remain in troubled waters at least until the election in 2001. However, the interesting question will be whether the future development will bring about new patterns of cooperation between the parties - perhaps even transgressing the traditional divide between "socialist" and "bourgeois" or left and right.

---

NOTES

1. The field work for the Norwegian Election Studies has been carried out by Statistics Norway (former Central Bureau of Statistics) since 1977. Up until 1993, the Election Study was a joint venture by Statistics Norway and the Institute for Social Research. Bernt Aardal and Henry Valen have been principal investigators for all these studies.

2. Among those who answered that the media campaign had a major impact on their vote, 56 percent decided on a party just prior to the election. Among those who said the campaign had some impact, 33 percent decided who to vote for just prior to the election, while among those who answered that they were not influenced by the campaign, only 16 percent decided just prior to the election.

3. The questions were formulated as follows: "Many people say it was very easy to decide which party to vote for in this election, while others found it very difficult. How did you feel about this; did you find it very easy, fairly easy, fairly difficult or very difficult?," "Looking back at the election this fall, would you say that you personally cared a great deal about which party won or lost the election, did you care to some degree, or didn't it matter to you personally?," "We would like to know whether you have taken part in any political discussions or conversations prior to this year's election. How frequently have you discussed the election with your family or with friends and acquaintances? Would you say it was daily, a couple of times a week, more seldom or never?"

ACKNOWLEDGMENTS

The author would like to thank Roy Ellis for help with the translation of this article.

REFERENCES

Aardal, B. 1998. "Stort/nyvalget i 1997 ble rekordenes valg," in Malmø, M. ed., Politisk
Årbok 1998 (Political Yearbook 1998). Oslo: Millennium.

Aardal, B. & Valen, H. 1989. Velgere, partier og politisk avstand. Oslo: Statistisk sentralbyrå.

Ansolabehere, S. & Iyengar, S. 1994. "Riding the Wave and Claiming Ownership Over Issues.
The Joint effect of Advertising and News Coverage in Campaigns," Public Opinion
Quarterly 58, 335-57.

Bjørklund, T. 1991. "Election Campaigns in Postwar Norway (1945-1989): From Party
to Media-Driven Campaigns," Scandinavian Political Studies 14, 279-302.

Christophersen, J. A. 1976. "Valget 1975. Tilbake til det normale?," Samtiden 1, 1-12.

Esaiasson, P. 1991. "120 Years of Swedish Election Campaigns. A Story of the Rise and
Decline of Political Parties and the Emergence of the Mass Media as Power Brokers,"
Scandinavian Political Studies 14, 261-78.

Iyengar, S. & Kinder, D. 1987. News That Matter. Chicago: The University of Chicago Press.

Martinussen, W. 1972. "Velgerne og de politiske stridsspørsmål 1965" in Valen, H. &
Martinussen, W. in cooperation with Converse, P. E. and Katz, D., Velgere og politiske
frontlinjer. Stemmegivning og stridsspørsmål 1957-1969. Oslo: Gyldendal Norsk Forlag.

McCombs, M. & Shaw, D. 1972. "The Agenda-Setting Function of Mass Media," Public
Opinion Quarterly 36.

Noelle-Neumann, E. 1974. "The spiral of Silence: A Theory of Public Opinion," Journal of
Communication 24, 43—51.

Noelle-Neumann, E. 1977. "Turbulences in the Climate of pinion: methodological
Applications of the spiral of Silence Theory," Public Opinion Quarterly 41, 143-58.

Noelle-Neumann, E. 1984. The spiral of Silence. Public Opinion - Our Social Skin. Chicago &
London: The University of Chicago Press.

Petrocik, J. R. 1996. "Issue Ownership in Presidential Elections, with a 1980 Case Study,"
American Journal of Political Science 40, 825-50.

Semetko, H. A. 1996. "The Media," in LeDuc, L., Niemi, R. & Norris, P., eds., Comparing
democracies. Elections and Voting in Global Perspective. London: Sage Publications.

Valen, H. 1976. "Norway: The Local Elections of September 1975," Scandinavian Political
Studies 11.

Valen, H. 1981. Valg og politikk - et samfunn i endring. Oslo: NKS-Forlaget.

Valen, H. 1984. "Hva skjedde ved valget?," in Valen, H. & Hanisch, T., eds., Likestilling,
arbeidsledighet og fordelingspolitikk. Noen resultater fra ettervalgundersøkelsen 1983,
Working Paper 1/84, Oslo: Institutt for samfunnsforskning.

Valen, H. & Aardal, B. 1983. Et valg i perspektiv, Samfunnsøkonomiske studier 54. Oslo:
Statistisk sentralbyrå.

Valen, H., Aardal, B. & Vogt, G. 1990. Endring og kontinutiet. Stortingsvalget 1989, Sosiale
og økonomiske studier 74. Oslo: Statistisk sentralbyrå.

Valen, H., Aardal, B. & Berglund, F. 1996. Velgerv■andringer og sviktende valgdeltakelse.
Kommune- ogfylkestingsvalget i 1995, Rapport 96:13. Oslo: Institutt for samfunnsforskning.